A Study And Analysis Of Portfolio Diversification
Definition of Diversification
Portfolio diversification is a widely recognized investment strategy that helps protect investors from the unpredictability of markets, as the old caution "Do not put all eggs in one basket" applies. As a risk management technique, diversification helps to surrender higher returns and set lower risks by mixing a variety of investments within a portfolio.
Consider an investor has an exposure to one asset class, such as to UK Equities. The earnings of the investor will fluctuate completely with that for UK Equities. But if the investor diversifies to also hold UK Gilts, some of the risks embedded in this portfolio can be removed without impacting on returns, since one performs strongly while the other may not. Take the equity market crash in the October 1987 as an example, UK Equities declined 27% for the month while UK Gilts increased by approximately 6%. In this situation, the security specific risks can be eliminated to a certain extent by gaining an exposure to other asset classes. (Frontier Capital Management, January 2009)
The key benefits of diversification are reducing portfolio loss and volatility, which are documented in both academics and practice. The Modern Portfolio Theory suggests that to obtain diversification benefits, the feasible correlation range should be between [-1, 1). By holding assets not perfectly correlated (ρ≠1), that is, do not move in perfectly same direction, the risks in a portfolio can be lowered and higher risk-adjusted returns can be achieved. In other words, the lower the correlation between assets, the greater the risk reduction can be obtained. Market Portfolio, a properly diversified portfolio reduces the all diversifiable risk to the level of non-diversifiable risk, through combination of all assets classes that generate the highest risk adjusted returns. This portfolio suggests diversifying as much as possible amongst uncorrelated assets, not to be restricted to one country but including all asset classes globally.
Related Literature Review
U.S. and other investors have already started investing in foreign securities markets. Global markets have become more incorporated; these may lead the broad tendency toward liberalization, deregulation in the money and the capital markets of developing countries. These modifications have resulted in greater correlation between national stock markets (decreased profits from international diversification). This may lead to future gains for the emerging markets such as Asia and Latin America. Nowadays the investors are looking to invest in emerging markets with promising international diversifications such as those of Central Europe.
According to the modern portfolio theory, the benefits of international portfolio diversification are connected with the correlations of security returns. The advantages of international diversification come up from the low correlations among developed and emerging equity markets have been confirmed in a number of studies such as Eun and Resnick (1984), Errunza and Padmanabhan(1988), Wheatly (1988), Meric and Meric (1989), Bailey and Stulz (1990), Divecha et al. (1992) and Michaud et al. (1996).
Early studies by Grubel (1968), Levy and Sarnat (1970) believed there are low correlations between index returns in different countries and argued the benefits of international diversification outweigh the huge costs, including transaction costs, regulatory and cultural differences, exchange rate risks. Since the foreign investments incline to be less closely correlated with domestic investments, diversification benefits could be obtained. For example, an economic downturn in the U.S. may not affect Japan's economy in the same way, which allows a U.S investor to have a small cushion of protection from Japanese investments against losses from the U.S. economic downturn. Recent evidence has shown that international diversification benefits are small for U.S. investors once transaction costs and short-sales constraints are incorporated (DeRoon, Nijman, and Werkerhenceforth, 2001). However, for investors in small, developing countries, global diversification may be much more important than for U.S. investors.
Washington University Theory suggests and results show that firm performance is at the beginning positive but eventually diminishes and becomes negative as international diversification enhances. The product diversification mitigates the connection between international diversification and performance. International diversification is negatively related to performance in non-diversified firms, positively related in highly product-diversified firms, and non-linearly related in sparingly product-diversified firms. International diversification is also positively related to intensity, but the interaction effects with product diversification are negative. The results of this study provide evidence of the importance of international diversification for competitive advantage but also suggest the complexities of implementing it to achieve these advantages.
A common feature of the above studies is that correlations between equity markets were estimated using relatively short-term horizons (weekly, monthly or quarterly). Kasa in 1992 mentioned that the benefits from international diversification affected by low correlations may be a misleading result for investors with long-term investment horizons if equity markets are trending together. Therefore, many current studies have used a combination of techniques to investigate whether there is any link and long-term actions between both developed and emerging equity markets. These examinations have created varied results and conclusions to the expenditure from diversification for US investors.
Investigating long-term connections between the US and European equity markets,
Kasa (1992) and Arshanapalli and Doukas (1993) generated confirmations of connections between the US with those markets; although, the results in Byers and Peel
(1993) and Kanas (1998) recommend that there is no such connection. Diversity in time
periods looked at **************** and research methods applied may explain the difference of outcomes between these studies. Studies of developed markets in the Pacific area have also created diverse findings. Campbell and Hamao (1992) argue that the US and Japanese markets are extremely incorporated, while results by Harvey (1991) and Chan et al. (1992) pointed out a lack of unification between the US and Asian markets. Sewell et al. (1996) confirms varying integration between Pacific Rim equity markets and the US.
DeFusco et al. (1996) report that the US market is not united with thirteen emerging capital markets in three geographical regions the Pacific Basin, Latin America and the Mediterranean. Felix et al. (1998) found no progress between the US and a number of developing markets. The noticeable autonomy of the US and emerging markets recommend the existence of long-term benefits from diversification across these countries.
However, several recent studies, such as those of Roll (1988), Hamao et al. (1990), Lau and McInish (1993), Rahman and Yung(1994), and Meric and Meric (1997), confirmed a significant rise in correlations and volatility transmission between equity markets during and after, the 1987 international equity market crash.
Empirical Analysis of Diversification
By conducting the empirical analysis, we'll study the effects of diversification into emerging market on the expected return and risk of investors' portfolios using efficient frontier and CAPM.
Efficient Frontier Analysis
We choose the Bond Index and Developed Market Index as the representatives of domestic market's risk and return level of U.S. investors, while using the Emerging Market Index as the representative of the emerging market. Our analysis is based on data from Oct.30, 1991 to Sept.30, 2010, totaling 230 samples (from DataStream). We have obtained the current U.S. 3-month Treasury bill rate as the risk-free rate (from Financial Time, annualized figure is 0.13%). To calculate the Efficient Frontier, we assume that investors can borrow and lend at the risk free rate of interest and that they are able to take short positions. By comparing the differences between the market portfolios and their efficient frontiers before and after diversification, we can conclude the influence of diversification into emerging markets on investors' efficient set.
First we conduct the empirical analysis on the whole time period: 1991-2010. By calculating the market portfolio comprised by Bond and Developed market index and the new market portfolio made by diversification into emerging market, we can compare their efficient frontiers and have an initial impression on the effects of diversification. The result is shown in graph 2-1.
Graph 2.1: 1991-2010
From the graph we can see that diversification into emerging market do bring us more expected return given the same risk than undiversified portfolio. But we also observe a slight improvement of 2.65% i.e. from 0.296 to 0.304 in the slope of efficient frontier. The risk-return improvement brought by diversification is quite insignificant.
Considering the large volatility of the correlation among assets and the riskless rate along with the time period, which may largely influence the effectiveness of our estimate on Market Portfolio and Efficient Frontier, we divide the 19-year data into 4 parts: 1991-1995, 1996-2000, 2001-2005 and 2006-2010. The following is the Efficient Frontier analysis of different periods:
Number of Assets
Graph 2.2: Period Analysis
(Note: Although the theta of three portfolio's efficient frontier in period 2001-2005 is -0.41, but the slope of the efficient frontier is positive. The improvement of the diversification into emerging market in this period is calculated by using this positive slope 0.41.)
As compared to the integrated analysis, the period analysis reveals a much more significant risk-reduction ability of diversification: the slopes of efficient frontier in the four periods are all largely improved, which is much bigger than that of the integrated analysis.
Graph 2.3: Correlation between Emerging and other two assets
The reason why the risk-reduction ability of diversification in smaller period is larger than that in the whole time range is that the effects of different return maximizing policies are offset by each other during a long period. We can see from the above chart that the efficient set can be achieved only by short selling the market portfolio and buying the riskless asset in the third period, while in the other three periods, the policies are opposite i.e. buying the market portfolio and riskless asset at the same time or taking short position on riskless asset in order to buy the market portfolio. These two opposite policies both have significant effects on maximizing returns and minimizing risk in their separate periods.
When conducting the integrated analysis, we are unable to separate these different effects making the influences of the third period offset by that of the others and showing relatively smaller and weaker risk-reduction ability than separate periods. But this insignificant figure can not deny the efficiency of diversification in reducing the risk and enhancing the returns. When we take a closer look at the correlation matrix (Graph 2.3), we find that the correlation between the Emerging and the other two assets (developed and bond) are very small which appear to be -0.04983 and -0.13109. According to the definition of Diversification we clarified above, the small negative correlations between these assets explain the significant risk-control effects by diversifying into the emerging market in the different periods. Therefore, the considerable influence of diversification on efficient frontier reflected in period analysis is really worth our attention.
Beta is a measure of the sensitivity of the asset's returns to market returns. We have obtained the betas for MSCI all country index , MSCI developed market index, MSCI emerging market index and Barclays US long government bond index which are 1, 0.98, 1.27 and -0.036.
As for the calculation of expected return of the three indexes by CAPM, we need to estimate three elements, namely risk free rate, beta of the index and market return. From the previous parts, we have obtained the betas of the three stocks and the market return which is the mean of MSCI all country index.
For MSCI all country index, the CAPM expected return is the original mean 0.00569, since it has a beta of 1. For MSCI developed market index, the beta is 0.98, which is very close to one, hence the CAPM generated expected return is 0.559, very close to the market return. For MSCI emerging market index with a beta of 1.27, higher than 1, which means there is a higher risk than the market thus, it has a CAPM expected return of 0.69. The Barclays US long government bond index has a negative beta of -0.036, which means it moves in the opposite direction of the market, hence, it has the lowest CAPM expected return of 0.1.
Recommendation and Limitation
Including the percentage by which expected return improves on average by efficient frontier.
Talking about benefits in terms of risk mentioning beta and correlation from capm.
Having looked at the advantages of diversification we would also talk about some of the disadvantages it possesses.
Trend is that markets are getting incorporated so influence of U.S all over the world. - Disadvantage- still worth investing;
Difficulties which range from unfamiliarity with infrastructure legal issues, accounting differences, and currency risk can have a dampening effect on international diversification. This article examines a group of chemical, electronic, pharmaceutical, textile, food, scientific, industrial, farm equipment and motor vehicle industries. Relate it last part of lack of information.*
Difficult to get detailed information about the asset, performance, market. Therefore difficult to decide based on assumptions and limited knowledge. This adds to the risk of your investment.
However having a thorough analysis of international diversification and looking both the merits and the demerits in detail it is evident that there are a number of benefits of diversification but some potential disadvantages as well. It can give you a lot of benefits if diversified properly that implies you have enough knowledge about the asset you are going to invest and the underlying trends of that assets together with its market potential which is hardly achieved by an ordinary person. Therefore it would be in the best interest of the investors to invest internationally if they have access to proper sources or professional advisors that have the appropriate information about the U.S market and the emerging market and the relationship between these two markets.
In the end we would talk about some of the limitations that were inevitable in our analysis.
1. Riskless rate- we use current Risk free rate though we have a data of 19 years; changes a lot in a long time period
2. Short sale- assume that we can short sale all the time but not possible in the real world.
3. Transaction costs- we ignore these costs however they do exist and whether we will be able to gain any benefits if they exist.
4. We only do our study on the basis of market instead of individual assets. First, market index may not really tell us the performance of the securities in this market. Second, since the investment of an investor can only include limited securities, the market based research can never best describe the effect of diversification in practice.
The CAPM is an important area of financial management. It is the linear relationship between the return required of an investment and its risk.
The CAPM is often criticized as being unrealistic because of the assumption, on which it is based. If the parameters of the model can't be estimated accurately then the definition of the market index is different or the company may have changed during the estimation period. Also if the model is right, there should be a linear relationship between returns and betas and the only variable which should explain returns, is betas. Unfortunately, the relationship between betas and returns is weak and other variables such as size, price, and value seem to have differences in returns.
Capm has a number of unrealistic assumptions
1-perfect capital market exists,i.e the market is efficient market (in equilibrium).
2-lending and borrowing can take place at risk free rates.
3-all investors have the same expectations about return and risk
4-capm works only when we r well diversified, only a diversied portfolio investor can use capm(unsystematic risk is not accounted for into capm)
5-risk is measured on the basis of historic returns patterns and assumption is that returns pattern will repeat in the future .
6-beta worked out from std. Dev. Of returns which are in turn measured on the basis of historic return pattern and also it is assumed that the pattern will repeat in future.
Use current Risk free rate; changes a lot in a long time period
Financial crisis; not easy to diversify