permanently attached to it, such as buildings. C.F. Sirmans et.al. 2003, in his paper observed that there has been a lot of efforts by researchers and academicians in studying and attempting to model the benefits of establishing diversification strategies for portfolio investments. All previous works concentrated on combination of different stocks into a single portfolio. With the passage of time research has been extended into analyzing individual factors (like bonds, currencies, real estate, international stocks); recently researchers did make some new in-roads in the potential benefits of real estate investment strategies.
In the same article the same authors cited the work of other researchers who have worked on Modern Portfolio Theory (MPT; see Markowitz, 1959), who have attempted to model the benefits of establishing diversification strategies for portfolio investments. Initial work only focused on potential gains from combining different stocks into a single portfolio. But research has been extended into bonds, currencies, real estate, international stocks and bonds.
According to Henderson Investors (2000), next frontier for the individual investor would be international real estate. Although microeconomic theory of consumer choice and decision-making under uncertainty is highly developed and finds much empirical support from stock market data, data providing similar evidence for direct, private investment in real property have not been easy to obtain. Accordingly, research in this area is scarce. Yet, many private investors prefer building a portfolio of real property investment, since it is believed that it offers a higher risk adjusted return than financial assets. The aim of the current research is to address the risk associated in real estate shared investment.
Over the years, many tools, processes and regulations have been created to give investors assurances that investment managers are weighing risks and evaluating potential returns in a fashion that is acceptable to investors. A large proportion of these mechanisms involve government regulation which
defines and enforces rules for regulated financial firms. Commercial banks, thrifts, broker-dealer firms, mutual fund companies, and insurance companies each have their own regulatory bodies and rules. Internally, these firms maintain staff to monitor and assure that the rules are obeyed.
Risk managers are charged with understanding, monitoring and controlling the new world of financial complexity. The definition of risk management has two separate parts. The first part defines it organizationally and the second part defines it functionally. Through widespread research on real estate investment trends, the risks associated with investing and the long term returns have been studied extensively (Springer, et al, 2005). It is essential that a potential investor has a thorough knowledge of “how to manage” the potential risks associated with an investment in real estate when analyzing the benefits of the potential investment. The specific risk factor variables can be tuned to suit the requirements and the particular real estate climate of the particular investment. The management of risks includes analysis of portfolio diversification, property age, specific demographics of the property locality, portfolio size and others.
Risks in real estate projects must be considered and should never be underestimated, because those tend to affect the whole project management processes, in terms of project programme delay, project cost overrun etc.
However, there is also a potential risk management technique, which can potentially affect the profitability of an investment: risk sharing. While the other risk management techniques focus on a single investing individual acquiring real estate, a group of partner investors may also choose to acquire real estate, with each having an interest in their investment. This concept of risk sharing changes the scenario as the burden of overall risk is split into the respective investment of each partner. Furthermore, this concept is different from the other concepts, especially if the dynamic nature of the real estate business is adapted in the modeling process because of the advantage of the shared contribution in the investment by various investing partners.
When a person acquires real estate, she/he also acquires a set of rights, including possession, control and transfer rights. Investment in real estate involves the commitment of funds to property with an aim to generate income through rental or lease and to achieve capital appreciation. However, the real estate income can be highly unpredictable and consequently investment in real estate is very risky as it is the case with investment in equity. One of the main reasons for the collapse of even the largest financial institutions, in the recent times, leading to the recession of major economies of the world is due to the collapse of the real estate business. This global crisis could have been prevented if the potential risk factors leading up to it had been identified at an early stage. But the risk assessment associated with this strategic investment was not properly followed. Thus, in order to be able to increase chances of investments in real estate returning profitable margins, it is critical that risk factors are identified and a proper investment strategy put into place.
Since real estate business involves immense risk owing to the requirement of mammoth funds devotion, the long operation period and the volatile markets, Brown (2004) attempts to explain as to why a potential investor would still consider investing in private real estate properties within a portfolio as opposed to conventional methods such as through the vehicles of REITs and stocks. He investigates the possible reasons for the private investor, especially in Tier II Properties (generally being apartment buildings between 4 – 100 dwelling units) as opposed to an investing institution or a pension fund. He concludes that (a) the private real estate market for “Tier II Properties” is inefficient and that (b) applying portfolio theory to individual parcels of real estate is, at best, a challenging task. At the practical level, high down payments, high transaction costs and lack of liquidity usually place forming a portfolio of private real estate assets out of the individual’s reach. On theoretical grounds, forming efficient set portfolios implies perfect liquidity, perfect divisibility and perfect reversibility. Even if the first two of these requirements are somehow met, the construction of a short sale of an individual parcel of real estate is impossible. Thus, a central value of the Markowitz (1952) model—being able to diversify away nonsystematic risk—is essentially unavailable to the private real estate investor.”
Furthermore, Brown (2004) states that the Tier II real estate market is defined to be privately owned investment real estate, holding a combination of ownership and control. Private real estate investing permits investors to influence the outcome. Hence, for these investors, probability is not purely random. In contrast to investing in other markets where investment is converted to financial assets, it is suggested that many prefer private real estate investment as they are more influenced by the dual factors of control and ownership. Mr. Brown also divided all other real estate two other tiers. Tier 1 is made up of smallest property containing 1 to 4 dwelling units, of which one is owner occupied. Tier 3 is the institutional properties where investment is converted to financial assets.
Roger Brown (2004) investigated the risk in real estate investment by conducting theoretical and empirical analysis of risk and returns accruing to individuals who were involved in real estate investments. Brown claimed that the returns are not normally distributed and that private real estate investors compensate for the distributional burdens their market imposes upon them by carefully assessing and controlling unavoidable non-systematic risk.
Brown’s work extends the work of Young and Graf (1996) by adding theoretical content, using different methods to generate return series, different technologies for estimating distribution parameters and investigating a real estate market with investors. Based on the concepts developed by Brown an attempt is made in this thesis to achieve the aim and objectives of the thesis as outlined below.
1.3 Aim and Objectives
The aim is to develop operating rules by constructing a dynamic risk sharing mathematical model that to be used by a private real estate investor who would like to share the risk with his/her partners and yet earn a good margin.
A tested dynamic risk sharing mathematical model to be used by the Real Estate partners in order to reduce risk.
A process that will be used to update various estimates involved in the model from time to time to capture the dynamics of the real estate market scenario.
1.4 Research Approach
Throughout the course of this thesis, an attempt will be made to evaluate numerous research articles providing in-depth analysis of these categories. With the added hindsight that past research work has uncovered in these fields, it is hoped to develop the most optimal conditions for a strategic real estate investment model.
In view of the above, an important consideration will be to enhance the core competencies in risk management.
The first step in this regard is to develop a risk-shared dynamic mathematical portfolio selection model for real estate business to be used by private real estate Tier II investors. Using this model, simulation of various alternative operating rules will be carried out.
Based on the outcome of the simulation, a strategy for implementation for the selected alternative will be developed. This will then lend immunity to any private investor against market instability, and other macro factors which influence the real estate market to a certain extent. This will help the investors who would like to share the risk with his partners as well as earn a good margin.
Real-life data will be collected and will be used to test and validate the model.
Last, strategies for implementation will be developed.
Chapter 1 presents the background of the chosen topic along with the aims and objectives, research approach
Chapter 2 presents a review of the existing literature detailing the hypotheses used and statistical and mathematical models used by the researchers in the past
Chapter 3 presents the methodology how the data will be collected which will be the back bone of dynamic risk-sharing mathematical model.
Chapter 4 presents the nature of the data collected and the methodology used for estimating the various parameters used in the mathematical model and test their sensitivity using appropriate hypothesis testing techniques.
Chapter 5 presents the operating decision rules derived by using appropriate non-linear mathematical programming technique.
Chapter 6 presents the sensitivity analysis of the parameters and the limits of the operating decision rules using simulation methodology.
Chapter 7 gives the strategies to be adopted for using the operating decision rules and the procedures for updating the parameters from time to time as required.
Finally, conclusions and future direction of studies are presented in Chapter 8.
Chapter 2: Literature Review
Investment can be considered as the employment of capital in the acquisition of real estate or interests within for undeviating ownership or for definite use of the person acquiring it. Real estate investments in general take up an enormous amount of funds; hence it is important to device a method in order to safe guard the interest of the investor.
Throughout the course of this thesis, an attempt will be made to evaluate numerous research articles providing in-depth analysis of these categories. With the added hindsight that past research work has uncovered in these fields, it is hoped to develop the most optimal conditions for a strategic real estate investment model applicable for Tier 2. Although a broad review of the research work related to portfolio design and management in connection with real estate business is presented in Chapter 2, a discussion of the salient issues related to real estate business is presented here to justify the need for carrying out further research in the area.
2.2 What is everything everywhere (EE) Model?
DiBartolomeo, et al (2005) have constructed a model based on a popular linear model for financial assets, the “Everything, Everywhere” (EE) model, which breaks discount rate risk into two components; the risk of treasury curve movements and the risk of changes in credit related yield spreads while linking global public security to over 50 factors. The principle behind this model have been applied to a real estate scenario, where the authors have identified that traditional real estate appraisals utilize one of three basic methods to value a property: (a) replacement cost, (b) comparable sales and (c) capitalizing the expected income. The proposed model estimates the risk and correlation at both the property and portfolio levels. This will then allow a potential investor to analyze the potential risk corresponding to a possible investment as well as the particular risky components. The authors believe that by assessing risk directly through the model, they allow the problems associated with traditional risk evaluations to be averted.
2.3 Influence of Global Real Estate Crash
Goetzmann et al (1995) analyze the risks of international real estate diversification, with a particular focus on the factors leading up to a global real estate crash. The authors suggest that a real estate market crash of a global scale has more to do with economic and monetary factors than the local factors of a market. Traditionally, there has been a much reduced risk associated with investment across international real estate markets leading to international investment generating high yield returns. However, there has also been evidence of great decline in some international markets which prove contrary to this, showing that this type of diversification has some drawbacks that the authors try to address and analyze.
2.4 Return Due to Diversification (RDD)
It is well established so far that due to the volatility in the real estate markets investing entails enduring a certain amount of risk. Both practitioners and theoreticians recommend holding a well-diversified portfolio to reduce risk.
Lee (2005) states that return due to diversification (RDD) effect makes the U.S. direct real estate a particularly attractive investments for long-term investors. However, he also adds that the results are dependent on the percentage allocation to direct real estate and the asset class replaced. Lee proposes that the addition of real estate into the mixed asset portfolio not just enhances the compound return of the portfolio but also reduces the risk. He tests this hypothesis by using the annual returns in the U.S. over the period 1951 to 2001 and finds that real estate can indeed justify a higher allocation in the mixed-asset portfolio than its individual compound return would suggest.
Lee next outlines the method of Booth and Fama (1992) for estimating the RDD of an investment within a portfolio. Modern portfolio theory shows that the lower the covariance of an asset with a portfolio, the higher its contribution to reducing the risk of the portfolio and so the greater the attractiveness of the investment to the portfolio. Booth and Fama (1992) illustrate that it is this insight that explains why the contribution of an asset to the compound return of a portfolio is greater than the weighted average of the individual compound returns.
Lee concludes his study report by establishing that the effect on portfolio RDD from an allocation to direct real estate is positive when it replaces large cap stocks – debatable if it replaces bonds and detrimental if it replaces small cap stocks. Thus, the argument for including direct real estate in the mixed-asset portfolio need not rest on its diversification benefits alone. A case can be made for adding direct real estate to the mixed-asset portfolio based on its contribution to portfolio RDD and so the compound return, or terminal wealth, of the fund from which the institution could meet its future obligations.
However, Lee does not differentiate between public real estate and private real estate investments and their separate inclusion in the portfolio. It may be interesting to find effect on portfolio RDD from an allocation to private real estate.
2.5 Diversification of Investment
Diversification has long been recognized as an effective portfolio management technique. It is based on the idea that spreading investment risk across a mix of diverse assets (i.e. whose returns are not correlated with one another) produces better risk-adjusted returns over the long term. This improvement in risk-adjusted returns is by virtue of negative returns from some of the assets being offset by positive returns elsewhere.
According to Kwame et al (2002), “diversification of investment (especially when it includes international investments) can open up a wider choice of investment opportunities, give improved risk-adjusted returns and reduce volatility when the investment is in real assets. Since the benefits of diversification are maximized when there is a low correlation between the assets, a well-diversified portfolio would include assets that are either negatively or lowly correlated.
However, it would appear that the investment issue, which remains absorbing and the focus of a lively debate in the finance literature, is the rationale for the superior performance of contrarian investment strategy. Williams (1995) was the first to make implicit reference to the contrarian investment strategy by hypothesizing and demonstrating through statistical simulation that ‘‘the greater the relative balance of return from operating and reversion, the more diversified the portfolio, and thus the better the portfolio performance.”
Kwame et al (2002) modified the model to conform to the Markowitz routine, and found that the association between the cash flow concentration level and the portfolio performance index, and that between the diversification index and the portfolio performance index were stronger than depicted by Williams (1995). This implies that diversification by sources of return could improve real estate portfolio performance.
The results on the returns of various indices indicate a high degree of variability and uncertainty. Especially Equity REIT’s index shows a coefficient of variation with an average return ranging from 99.24% to 326.25%.
Kwame et al (2002) study, in conclusion, verifies Williams’ hypothesis that diversification by sources of return could complement the traditional diversification strategies to significantly improve real estate portfolio return.
Thus this review lends to the aim which suggests that diversification, in fact, could improve real estate portfolio returns, however what investment alternatives or parts thereof exactly ought to be the composition of the portfolio in order to optimize returns has not been elaborated on. That is, the degree of diversification and the extent to which inclusion of investment alternatives or parts thereof do not get mention in the study.
2.6 Contribution and Optimal Levels of Inclusion of Different Investments
The basic objective in developing a product portfolio is to maximize its return with minimum risk. In order to develop a portfolio with a low standard deviation signifying lower levels of risks, one needs to be familiar with the past development on the subject.
Based on a 25 year observation of direct private real estate (through NCRIEF equity index) and a 30 year observation for public real estate (through NAREIT equity Index) in assessing the return contribution in a mixed asset portfolio, Mueller and Mueller (2003) explore the contribution and optimal levels of inclusion of different investments in a return maximizing portfolio. The five time periods analyzed are the 5,10,15,20 and 25 year annual returns.
Substantial fieldwork has been accomplished in order to provide a foundation for the research paper.
Gilberto’s (1990) comparison of public and private real estate returns.
Mueller, Pauley and Morrill’s (1994) inclusion of REITs in a mixed-asset portfolio.
Miles and Tolleson’s (1997) revision of different public and private debt and equity investment alternatives.
14.Ziering and McIntosh’s (1997) study of the benefits of including both REITs and core real estate (using NCREIF returns) in a mixed-asset portfolio of stocks and bonds from 1972 to 1995.
Gordon, Cantor and Webb’s (1998) studies of the portfolio diversification effects of international real estate securities on a mixed-asset portfolio of U.S. stocks, corporate bonds, real estate securities and international common stocks.
Chua (1999) studies on the role of international real estate in a mixed-asset portfolio while attempting to control for higher taxes, transaction costs and asset management fees incurred when investing in real estate, as well as the appraisal smoothing in real estate return indices.
Gilberto, Foort, Hoesli and MacGregor’s (1999) test of the predictive powers of an optimal diversification strategy within a mixed-asset portfolio using a threshold autoregressive conditional heteroskedasticity model (QTARCH).
The above mentioned studies along with the findings of Ling and Naranjo (1999), Quan and Titman (1999), Ziering, Liang and McIntosh (1999), 19.Fu and NG (2001), Ciochetti, Craft and Shilling (2002) and Feldman (2003) had been critically analyzed in order to come to a viable conclusion. The analysis took into account returns, volatility, correlations and Markowitz efficiency frontier.
2.7 Real Estate Investment Trust (REIT)
Unlike real estate directly held by the investor, REITs are a liquid asset that can be sold fairly quickly to raise cash or take advantage of other investment opportunities. Using REITs, investors can diversify their holdings between various geographic areas and property specializations. REITs can tap the debt and equity markets and raise funds to take advantage of opportunities when they arise. REITs have a lower correlation to equities than many other asset classes, providing portfolio stability for those with an active asset allocation strategy. Serrano and Hoesli (2007) analyze the part played by financial assets, direct real estate, and the Fama and French (1993) factors in amplifying equity real estate investment trust (EREIT) returns and scrutinizes the expediency of these variables in forecasting returns. His study recognizes the assertion that equity REITs (EREITs) are investments whose fundamental assets are stocks, bonds, and real estate; thus uses aggregate substitutes for the set of economic and financial variables that would be useful in forecasting EREIT returns. Hence, the study by Serrano et al examines the leeway of making lucrative forecasts based on the conclusions that securitized real estate is a hybrid asset.
In order to determine the suitable model for making useful predications, Serrano and Hoesli (1993) use four models:
Capital Asset Pricing Model (CAPM) of Sharpe (1964);
CAPM with the Fama and French (1993) Factors;
Clayton and MacKinnon (2003) Hybrid Model; and
Clayton and MacKinnon (2003) Model with the Fama and French (1993) Factors
The study then examines the forecasting ability of the four securitized real estate return-generating models by employing three forecasting methods i.e. Time varying coefficient (TVC) regressions, Vector autoregressive (VAR) systems, and Neural networks models.
Forecasting accuracy has been measured with traditional statistical criteria, as well as by comparing active investment strategies based on the papers forecasts to a passive buy and- hold strategy. This helps determine not only which model specification is the most appropriate for securitized real estate forecasting, but also which forecasting technique makes the most accurate predictions. The methods used are Mean Error, Root Mean Squared Error, Mean Absolute Error, Directional Accuracy and Theil’s U2 Inequality Coefficient. In order to empirically validate the above mentioned measurement techniques, the researchers obtained data from Thomson Datastream except for the real estate series and the Fama and French (1993) factors. All indices used are quarterly total return indices for the period 1978– 2006. For securitized real estate, the FTSE NAREIT EREIT series is chosen. Datastream’s total market index is used for stocks, and the Merrill Lynch’s 7–10 year government bond index is used for bonds. As a risk-free rate, the Euro-Currency three-month middle rate is retained. The size and book-to-market factors have been provided by Kenneth French. Finally, the NCREIF Property Index (NPI) is used for direct real estate. Real estate returns are unsmoothed using the approach proposed by Geltner (1993).
The results of the above study recommended that EREIT returns are optimistically correlated to stock, size, and book-to-market factors. None the less, these associations are volatile, with stocks and size being overriding until the early 1990s, while the book-to-market and size factors dictate thereafter. With bonds, a usually positive but frail liaison is found, whereas with real estate, the relationship demonstrates much unpredictability and seems to be recurring. This study highlights the significance of models including the Fama and French (1993) factors, as well as the superiority of neural networks as a forecasting tool.
In particular, the hybrid nature of real estate securities can be exploited for prediction purposes, although supplemented with the aspect of optimizing Risk Adjusted value of total returns from portfolios could be a significant inclusion in the study negated by the authors.
Waggle et al (2006) concentrate on determining the minor effects of alterations; due to non-stationarity or assessment errors in the REIT-stock risk premium and the REIT-stock correlation on the most advantageous portfolio asset mix of REITs, stocks, and bonds. The study also uses historical return data for REITs, stocks, and bonds to generate base level assumptions.
“Monthly and annual total return data for equity REITs, large-company stocks, and long-term government bonds for the 1972 to 2002 period is primarily focused upon.
The marginal effects calculations described begin with assumed values based on the 1988 to 2002 period, the standard deviations of REITs, stocks, and bonds are assumed to be 15.8%, 18.6%, and 11.2%, respectively, The REIT-stock correlation is 0.36 and the REIT-bond correlation is 0.14, while the stock-bond correlation is 0.13. These assumptions have been taken constant throughout the paper.
On the other hand the assumed differences in the returns between REITs and stocks and between stocks and bonds vary through their analysis. In this study the functional decisions are based on the supposition that investors choose the portfolio weights that capitalize on utility U with the common function, given in the paper as:
where, higher values for A equate to higher degrees of risk aversion and vice versa. Levels of A ranging from 1 to 10 are examined. The study goes on to calculate portfolio return for the three asset case with REITS, stocks and bonds, and then the portfolio variance. It assumes two constraints to rule out portfolio short selling and assure portfolio completeness. Further the paper calculates the optimal portfolio weight, and then the calculated marginal effects of change on the portfolio weights are analyzed. The marginal effects due to changes in returns are exaggerated by the variances and covariances of the asset returns and the level of risk aversion of the individual investor, but not by the current returns levels.
The findings signify that the expected return of REITs relative to that of stocks is a much more imperative factor than the REIT-stock correlation in making portfolio decisions. The portfolio impacts due to variation in REIT returns are more evident for aggressive investors and less for more conservative investors. For many investors, the marginal effects calculated in this paper revealed that their actual, as compared to theoretical optimal, allocations would not be affected at all.
Cite This Work
To export a reference to this article please select a referencing stye below: