REAE 5311 Blog Post
It is well known in finance that diversification reduces risk, and if this diversification is done internationally, more risk sources could be offset. However, and for the particular case of real estate, international diversification is not only acquiring assets abroad: it is a strategic allocation process in which investors optimize their investment yield while significantly reducing risk. In order to prove this, there have been several empirical research papers that precisely address this issue. The purpose of this blog post is to present some of the findings of these papers and to stretch the importance of a good international diversification.
Even though real estate literature is not as wide as with general finance, this topic has enough breadth to cover several interesting points. The first paper analyzes the performance of publicly traded real estate companies in international markets, having these companies as a proxy for real estate markets. The following paper focuses exclusively in emerging markets, as it analyzes several property indices in these countries. Then, the third paper addresses a more complex model: Cointegration among markets. Finally, there’s a paper which attempts to demonstrate the relationship between fundamental and market prices on real estate properties. This selection might seem rather limited; nevertheless, it addresses some of the most important issues in this theme.
Publicly traded real estate companies
In the paper Commercial Real Estate Return Performance: A Cross Country Analysis, by Ling and Naranjo (2002), the authors address the importance of using publicly traded real estate companies with a very intuitive rationale: since it is virtually impossible for investors worldwide to hold and manage real estate assets in foreign countries, investments in public real estate companies are considered as a proxy for owning actual real estate. Therefore, in order to have international real estate diversification, an investor should focus on stocks from these companies.
First, the authors measured the growth of the stock price of real estate companies in 30 countries, along with several profitability ratios. They also calculated betas, the measure of sensibility of the stock of the real estate company compared with a world wealth portfolio. These betas were later used on regressions to find out whether they absolutely account for the movement of the price of the stock. If they do so, we could conclude that there is no need for international diversification. Conversely, if the residuals of the regressions also have explanatory power in returns, we could also conclude that there are country specific factors that influence the return on the stocks. The latter was the hypothesis the authors found more accurate.
The authors performed several other robustness checks for their findings, and concluded that the best strategy for international diversification was not picking “winners”, but rather to focus on the covariance structure of the assets. This is consistent with portfolio theory: covariances are the ones that really determine the amount of diversified risk. They also concluded that there’s a worldwide factor that affects real estate returns, however, country-specific factors also play an important role in determining the returns of a stock. Therefore, diversification is necessary in order to reduce these local factors.
Levent, Istanbul, Turkey
Real Estate in emerging markets
Barry and Rodriguez, in their 2004 paper Risk and return characteristics of property indices in emerging markets, focused on analyzing real estate in 15 emerging markets. The first major hurdle they encountered in their analysis was exchange rate: returns in local currency, especially in emerging markets, will be higher than in US dollar terms. For example, currencies in Brazil, Turkey, Argentina, and Venezuela suffered major depreciations in the last ten years, making returns in local currency not representative for international investors. Hence, it is important to change everything to a single currency, in this case the US dollar, in order to have a more accurate rate of return for a foreign investor.
After addressing this issue, the authors compared the world and emerging market broad market indices against their respective property indices. There is a big difference in performance between both emerging market indices; nevertheless, when examining specific country examples, we can see that this underperformance is driven by specific countries. There is also another issue to examine: total risk. Returns should be adjusted to risk in order to be comparable and therefore determine the profitability of a market. For this task, the authors calculated Shapre Ratios, the ratio between returns and standard deviation, for the sample countries. After these adjustments, the authors found out that there were countries that were not significantly influenced by global capital markets, and, in spite of the fact that the overall performance of real estate indices was rather poor, the risk adjusted performance significantly improves the scenario.
The authors then suggested, based on these findings, that there could be diversification potential by investing in developing real estate markets. However, some of these investments could be troublesome: excessive regulation and low market capitalization could offset the potential benefits of diversifying. There is an exception to this issue: unsecuritized investments during times of crises. However, it is important to have data relative to property type and geographic areas within countries in order to have a really accurate investment decision. Unfortunately, these data is not available, therefore making this analysis a very difficult goal to accomplish in the short run.
Sao Paulo, Brazil
Cointegration
Modeling Linkages between US and Asia-Pacific Securitized Property Markets, a paper from Yunus and Swanson (2007), utilizes the technique of Cointegration in order to determine whether there are long run common factors among markets. It is first important to define Cointegration as a process in which two or more time series share similar characteristics during a certain time, therefore not allowing for diversification benefits. What this paper analyzed was if there were Cointegration relationships among the sample countries (US, Singapore, Australia, Hong Kong, and Japan), and some other short-run relationships.
It is important to note that diversification benefits could be time-specific, and if there are structural breaks in the data, subsamples should be taken into account. For these specific countries, there is a structural break sometime between 2002 and 2003. The authors then decided to divide their sample period, having year end 2002 as their period differentiator. This date is intuitively accurate: it marks the beginning of the boom in commercial real estate markets. It is also important to note the fact that all returns were converted to US dollars, in order to be consistent among returns and therefore be able to have accurate comparisons.
After analyzing long-run Cointegration relationships, the authors concluded that the United States, Australia, and Singapore are cointegrated, therefore reducing any potential benefits from diversification. On the other hand, Japan and Hong Kong could have a better diversification potential, since those markets are not trending with the other three markets. There is an intuitive explanation for this finding: the markets in the United States and Australia are the oldest from the sample, and Singapore modeled its property laws based on Australia. On the other hand, Japan and Hong Kong might have also suffered from consequences due to a GDP drop during that same sample period.
On the other hand, short-run causality tests were also performed for the same countries. Since this study was performed with daily data, time lags were included to be consistent with a cause-effect relationship and market operation hours. In other words, Asian shocks on day t were compared against American returns on day t, but American shocks on day t were compared against Asian returns on day t + 1. After the analysis, the authors found no lead-lag relationships, basically stating that all markets are short-term independent. This is intuitively sound: real estate is most of the time considered as a long term investment.
Hong Kong
Fundamental Prices
Finally, Hott and Monnin (2008), with their paper entitled Fundamental Real Estate Prices: An Empirical Estimation with International Data, give a theoretical framework for determining fundamental prices for real estate. This theoretical framework is very important, since it could be a good way for identifying market bubbles. Whenever market prices significantly deviate from their fundamental prices, we’re on a presence of a disequilibrium situation, and the market will eventually take care of it. This situation is rather self explanatory with the American real estate market during the last year.
In order to calculate the fundamental price of a property, the authors considered first calculated what is called the imputed rent. This rent is influenced by three factors. The first one is the mortgage rate multiplied by the price of the property, in other words, the interest payment for the property. Next, there is a fraction of the value of the house that is paid as maintenance cost. Finally, the expected capital gain, as the difference between the net fiscal effect of the expected price of the house one period ahead minus the price in the current period. Rearranging for the price of the house, the authors found it was a function of the imputed rent, the mortgage rate, and the expectation of the price of the property for the following period.
The authors then used two approaches for interpreting the evolution of imputed rents: either a no arbitrage condition, represented by the imputed rent approach, or a market equilibrium condition based on supply and demand (called S/D model by the authors). The authors then used these two approaches against data from USA, UK, Japan, Switzerland, and the Netherlands. The results were somehow as expected: market values were most of the times different than fundamental values represented by the two models.
The authors conclude that there is no clear evidence on whether market prices converge with fundamentals in the long run. However, the gap between fundamental and market prices is slowly decreasing, suggesting that prices might converge in the future. There is also a possibility of forecasting prices for properties by using this technique: the authors demonstrate that their models were accurate on determining the actual movement of the market. There is no explanation on why does this happens, but the authors indeed demonstrated that there is a difference between fundamental and market prices.
Canary Wharf, London, EnglandWhat’s next?
Empirical literature in real estate is still growing, and there are several issues that haven’t been completely addressed, and empirical research should be moving towards that direction in the future. The first one is regional cycles. Differentiating among countries, especially developed, is not a difficult task. However, when including developed countries, the action becomes more difficult. Then, regions within countries would have to be the next sub division. Since real estate investments require a large amount of capital and a long term horizon, the allocation of a specific region within a country becomes important. Even though there are intuitive reasons to move towards a specific place, empirical research should be made and models should be proposed to effectively propose a regional diversification within countries.
A second proposal is the usage of property types. This is somehow analogous to industry diversification in the stock market. It is not the same to invest in industrial property as opposed to houses. Property types could mark an important difference for allocating resources. Then, adding property types to the international scenario, this makes things more complicated, basically due to the lack of data. However, there should be some other proxies used in order to fill in this gap. Intuition could give us a good direction, but there should be a model that could give an investor a better idea on what to do.
A third proposal is monitoring real estate through the inclusion of a “standard deviation” variable. What I mean with this is that, since the real estate market is not perfect, the difference in prices of real property might give more information to potential investors than prices themselves. Property indices are good, but the market will need a more dynamic indicator in order to have accurate information for investors. This might require a structural change in data gathering, but probably a theoretical model explaining this need will lead to a more comprehensive data set.
Finally, with these three new sets of data, more short and long run relationship tests will also be needed. The Cointegration technique, among others, should still be used to find similar trends in time series movements and therefore maximize the potential for diversification. On the other hand, regulation should also be analyzed, since there’s evidence that when countries engage in similar legal frameworks, some benefits from international diversification are offset. Finally, countries should create financial instruments similar to real estate investment trusts (REIT’s), in order to offer international investors a better possibility for real estate investment.
Conclusion
Empirical research in real estate has been increasing in the last years. The purpose of this post was to analyze four papers in different areas of real estate, and then give some ideas of the future movement of real estate empirical research. The papers are summarized as follows:
1.- By analyzing real estate publicly traded companies in several countries in the world, there is evidence of the existence of country specific risk. This suggests that international diversification could offset this local source of uncertainty.
2.- Emerging markets offer attractive diversification possibilities, and even though the emerging market property index seemed to underperform the world financial market, investing in specific country actually give better diversification results .
3.- There are long-term Cointegration relationships between the United States, Australia, and Singapore. Investing in Japan or Hong Kong could increase the potential of diversification for international investors. Markets are independent in the short run, consistent with the fact that real estate is a long term investment.
4.- Fundamental prices of real estate are a function of rents and of market equilibrium models. When compared to market prices, there is an important gap. However, there is evidence that prices seem to go back to fundamental values in the long run.
Finally, it is important to say that there are still several possibilities for empirical research in real estate. However, the fundamental principle is simple: international diversification in real estate is a strategic asset allocation activity that deserves to be fully analyzed in order to obtain the highest possible benefits.
Bibliography
- Barry, Christopher B. and Mauricio Rodriguez, 2004, Risk and return characteristics of property indices in emerging markets, Emerging Markets Review 5, 131-159.
- Hott, Christian and Pierre Monnin, 2008, Fundamental real estate prices: an empirical estimation with international data, Journal of Real Estate and Financial Economics 36, 427-450.
- Ling, David C. and Andy Naranjo, 2002, Commercial real estate return performance: a cross-country analysis, Journal of Real Estate Finance and Economics 24:1/2, 119-142.
- Yunus, Nafeesa and Peggy E. Swanson, 2007, Modeling linkages between US and Asia-Pacific securitized property markets, Journal of Property Research 24:2, 92-122.
Pictures
1 comment:
Fantastic blog! I actually love how it's simple on my eyes as well as the data is well written. I am wondering how I may be notified whenever a new post has been made. I have subscribed to your rss feed which really should do the trick! Have a nice day!
Estate Agents Canary Wharf
Post a Comment