Idzorek et al.’s first caveat is that the two forward-looking asset allocations sets they present represent only two of the many possible asset allocations that can be derived from analytically based forward-looking expected returns.
Second, it may questionable to use global REITs and stocks of real estate companies to embody the long-term performance of the universe of commercial real estate equity investments. However, considering REITs and listed real estate stock returns as a proxy for all commercial real estate investments has become more suitable in the last few years as these investments have been growing fast and represent a larger proportion of the market. In the end, we are confronted with an empirical question.
Third, investors who have a separate strategic asset allocation to REITs and listed real estate stocks may not need to own these indirect real estate investments. However, all other investors should own REITs and listed real estate stocks. Idzorek et al. note that this caveat is relevant only to a small group of the largest investors.
Fourth, the CAPM-based asset allocations presented are market-based and assume that investors do not suffer from a U.S. or home bias (i.e. the tendency to invest in a large proportion of domestic assets, despite the supposed benefits of diversifying into foreign assets). Another definition of the market portfolio (one that includes, for example, commercial real estate) will produce another asset allocation. More work on this topic is needed.
Fifth, if investors would be able to increase their opportunity set so as to include all the most important assets that are part of the – unobservable – market portfolio (assets such
as TIPS, convertible bonds, commodities, emerging market bonds, high-yield bonds and emerging market stocks), the end result will be a reduction in the allocation to the asset classes considered in Idzorek et al.’s study.
The sixth caveat, which is related to the fifth, is that a different composition of the market portfolio will yield a different set of Black-Litterman-based asset allocations.
Finally, actual asset allocations should be tailored to the investor’s unique situation.
Example:
Let us use these insights on real estate asset allocation in a hypothetical example. Peter Gray, an investment analyst at YHG Investments, is analyzing whether to include real estate as an asset class in a client portfolio that is currently constituted only in U.S. stocks and U.S. bonds. To that end, he collects the following historical information on U.S. real estate, U.S. stocks, U.S. bonds and U.S. inflation.
YEAR REITS Stocks Bonds CPI
1991 35.68% 30.47% 16.00% 2.98%
1992 12.18% 7.62% 7.40% 2.97%
1993 18.55% 10.08% 9.75% 2.81%
1994 0.81% 1.32% -2.92% 2.60%
1995 18.31% 37.58% 18.47% 2.53%
1996 35.75% 22.96% 3.63% 3.38%
1997 18.86% 33.37% 9.65% 1.70%
1998 -18.82% 28.58% 8.69% 1.61%
1999 -6.48% 21.04% -0.82% 2.68%
2000 25.89% -9.10% 11.63% 3.44%
2001 15.50% -11.89% 8.44% 1.60%
2002 5.22% -22.10% 10.26% 2.48%
2003 38.47% 28.68% 4.10% 1.87%
2004 30.41% 10.88% 4.34% 3.29%
2005 8.29% 4.91% 2.43% 3.40%
2006 34.35% 15.79% 4.33% 2.53%
Average 17.06% 13.14% 7.21% 2.62%
St. Deviation 16.46% 17.45% 5.64% 0.64%
CORRELATION MATRIX
REITS Stocks Bonds CPI
REITS 1.00 0.17 0.22 0.31
Stocks 1.00 0.18 -0.19
Bonds 1.00 -0.09
CPI 1.00
Note: “REITS” returns were measured from the total return index REITs calculated by NAREIT, “Stock” returns were measured from the Standard and Poor’s 500 total return index,
“Bonds” returns where measured from the Lehman Aggregate U.S. Bond total return index, and “CPI” is the percentage changes in the U.S. Consumer Price Index.
Part A: Peter notices that real estate provided the highest return of the three asset classes, but he is concerned that the standard deviation of real estate is almost as high as the standard deviation of stocks. This leads him to question whether to add real estate to the traditional portfolio at all. Do you agree with Peter?
Modern portfolio theory suggests that asset classes such as real estate, which have low correlations with the current opportunity set of traditional asset classes (we can see in the table that it has correlations of 0.17 and 0.22 with stocks and bonds, respectively), may well provide substantial diversification benefits. Therefore, even though real estate exhibited one of the highest standard deviations, its low correlation to traditional assets, combined with the high returns it offered, makes this alternative investment a good potential candidate to be added to a traditional portfolio.
Part B: Based on the information presented in the tables, Peter is assessing whether real estate offers a better hedge than stocks or bonds against the risk of inflation. What would you say to him?
You could say that real estate investments exhibited a positive correlation with inflation (0.31), as opposed to stocks and bonds, which showed a slightly negative correlation with inflation (-0.19 and -0.09, respectively). Therefore, it appears that real estate investments offered a better hedge against the risk of inflation than stocks or bonds did during the period 1991-2006. A more rigorous study would attempt to see the correlation of real estate returns with inflation using a measure of unexpected inflation (what is shown in the table, CPI inflation, is simply the observed inflation, whether it was expected or not).
References
Perold, A. F., and W.F. Sharpe. “Dynamic Strategies for Asset Allocation.” Financial Analysts Journal. January/February 1988.
Chhabra, A. “Beyond Markowitz: A Comprehensive Wealth Allocation Framework for Individual Investors.” The Journal of Wealth Management, Spring 2005.
Erb. C., and C. Harvey. “The Strategic and Tactical Value of Commodity Futures.” Financial Analysts Journal. Vol. 62, No. 2, March/April 2006.
Idzorek, T.M., M. Barad, and S.L. Meier. “Global Commercial Real Estate.” The Journal of Portfolio Management. Special Issue, 2007
9
Current Topics
M a i n P o i n t s
Explaining what happened in the Amaranth debacle, how it happened and how it could have been avoided
Applying the market events of August 2007 to the concept of systemic risk Describing factors contributing to the subprime credit crisis and
recommending policies and practices to address them