DESIGNING LOWER VOLATILITY PORTFOLIOS
By Lou Stanasolovich

Part 1 of a Series

Since the inception of Risk-Controlled Investing, we have received many compliments and a number of requests have been made to add new features. One such request was, how does one go about designing a Lower Volatility Portfolio? As a result, we are creating a column for the newsletter simply called “Designing Lower Volatility Portfolios.”

In this issue, we will design two fairly simple base portfolios upon which other investments can be added and therefore risk as well as return can be added and/or subtracted away depending upon the circumstances. When constructing any investment portfolio, it is important to determine its diversification strategy. The logical process is to determine the correlation between each investment. 

Redefining the Correlation Coefficient:
In our last issue of Risk-Controlled Investing, we wrote an article about understanding the correlation coefficient, and its importance in being a key element in the construction of a Lower Volatility Portfolio. This calculation captures the similarity or dis-similarity in the price movements of two securities. A correlation of 1.0 implies that two securities move, to the same degree, in the same direction. A correlation of -1.0 implies that two securities move, to the same degree, in the opposite direction. Realistically, most pairs of securities lie somewhere between -1.0 and 1.0. When searching for additional investment opportunities for a Lower Volatility Portfolio, discovering securities that historically have had low correlations relative to the existing investments within the portfolio are essential in the continuation of its low risk characteristics. Following this philosophy can help to achieve the level of diversification needed to seek capital preservation while allowing the potential for capital appreciation.

What is Diversification:
Diversification represents the idea of not putting all assets into one particular investment. Well, how do investors avoid this potentially detrimental scenario? A simple solution is to spread the assets out over multiple asset classes. By doing so, some risk of the portfolio can be eliminated. Non-systematic risk, which represents risk unique to a specific investment only, not the overall market in general, becomes minimized. However, systematic risk, defined as market risk cannot be eliminated by continuing to add additional securities to the portfolio. Therefore, true diversification lies among the correlation between the underlying securities and/or asset classes of the portfolio.

Legend Financial Advisors, Inc.® and now Risk-Controlled Investing have been experimenting with model portfolios for a number of years to develop Lower Volatility Portfolios. In doing so, we have constructed an Experimental Model Portfolio (#1) utilizing the following asset classes with their respective allocations. 

Floating Rate Senior Secured Bank Loans:
The Eaton Vance Senior Floating Rate Fund (EIFRX) represents the fixed income portion of the portfolio. Unlike the traditional fixed income investments in bonds which fall in value as interest rates rise, the fund invests in senior floating rate bank loans [the interest rate on the underlying loans adjusts upward or downward in the same direction as the Prime Rate or LIBOR (the London Bank Inter Offer Rate – this is essentially an international prime rate) does], which are typically below investment grade and benefit from higher interest rates. Therefore, the fund can provide a high level of current income. Prior to the inception of the fund in May of 1999, returns were utilized from the Eaton Vance Prime Rate Reserve Fund. Monthly returns have historically ranged from -0.79% to 1.08%. The fund does not display much volatility. In fact, the fund is utilized to provide stability in the portfolio. The 10 year compound return is 5.16% with a 10 year standard deviation of only 1.10%.

Real Estate:
The Cohen and Steers Realty Shares Fund (CSRSX), a mutual fund of REITs, represents the Real Estate allocation. Managers Martin Cohen and Bob Steers pursue a macroeconomic overview of the economy and regional markets, in their search to discover growth potential among their stock selections. The fund has produced a ten year compound return of 15.81% with a ten year standard deviation of 14.42%. 

Commodities:
The PIMCO Commodity Real Return Fund Institutional Share Class (PCRIX) represents the commodity portion of the portfolio. The fund’s return is designed to track closely the Dow Jones/AIG Commodity Index. Prior to the fund inception in July, 2002, the returns have been utilized from the Dow Jones Commodity/AIG Commodity Index with the PIMCO fund’s expense ratio being applied to net the returns. The ten year compound return is 14.09% with a standard deviation of 16.96%.

Standard & Poors 500 Index:
A capitalization weighted S&P 500 Index has been utilized as the fourth asset class. During the ten year time frame, the compound return is 8.96% with a standard deviation of 17.54%.

The strategy behind the construction of this portfolio was to define the relationship between asset classes that when standing alone can be extremely volatile. With the exception of Bank Loans, each asset class has a high standard deviation due to the high risk characteristics. So, what is the point here? Four asset classes have been equally-weighted in a combined portfolio. Does this mean that the portfolio will maintain a high risk profile? First, view the correlation matrix below displaying the correlations among all four asset classes. As the Correlation Matrix displays, each asset class has a low if not negative correlation with each other, which as previously mentioned, is essential in the construction of a Lower Volatility Portfolio. 





The following table and charts provide the results of Experimental Portfolio #1, which has been back-tested since January, 1996. (Please note this Experimental Portfolio is not a recommended or actual portfolio at this point in time. We are merely attempting to illustrate basic investment concepts in constructing a Lower Volatility Portfolio. Also, because this is an Experimental Portfolio, we are not reducing the returns by our management fees.) As you can see Experimental Portfolio #1 outperforms the S&P 500 by 294 basis points per year yet exhibits less than one-half of the volatility. Experimental Portfolio #1 produces negative monthly returns only 28.95% of the time while the S&P 500 is negative 38.60% of the time. The Reward to Risk Ratio (The Standard Deviation divided by the Annualized Compound Return) is 1.406 for Experimental Portfolio versus .511 for the S&P 500, which approximates its long term number as well.

We have also developed Experimental Portfolio #2, attempting to decrease the standard deviation even more. By simply allocating a double weighting to the Fixed Income portion of the portfolio, the following results have been obtained:










The effect of applying a double weighting to the fixed income allocation resulted in only a 1.31% reduction in the compound return with a 1.70% reduction in the annualized standard deviation. 

The purpose of the experimental portfolios was to capture the benefit of a truly diversified investment portfolio. It was not done by developing a 5, 10, 15, 20, etc. security portfolio. It was done by combining four asset classes, each of which have low correlations to each other, in an equally-weighted portfolio. By applying this strategy, investors can weather bear markets as opposed to a portfolio comprised of highly correlated assets. None of the asset classes have a correlation higher than 0.37 (Real Estate vs. S&P 500). As a result, when the stock market is experiencing downward pressure, a portion of the portfolio should be experiencing positive returns. 

One caveat that should be stated for the S&P 500 as well as both Experimental Portfolios #’s 1 and 2 is that returns are unlikely to be as good over the next decade as they were from 1996 through the first half of 2005. With the S&P 500 over-valued by 30% to 40% over its historical normalized ratio and REITs 15% to 25% over their normal valuation measures, it would appear that returns moving forward will not be as robust as they were in the past. One factor though in favor of the experimental portfolios is the investment opportunities going forward that both bank loan funds and commodities provide. Rising interest rates will obviously favor bank-loan funds. Commodities despite their strong returns over the last few years appear to be only in the fifth inning of a nine inning game. That is what true diversification is all about.

Legend Financial Advisors, Inc.
5700 Corporate Drive, Suite 350
Pittsburgh, PA 15237-5829
Phone: (412) 635-9210
Fax: (412) 635-9213
Toll Free: (888) 236-5960
E-mail: legend@legend-financial.com
Web Site: www.legend-financial.com