Quasi-Commodities Anyone?

by Craig L. Israelsen
Reprinted from Financial Planning Magazine
July 2007

dding "alternative" assets to investment portfolios is gaining increased acceptance, or at least, increased consideration. To be sure, the definition of alternative assets differs from person to person. In this analysis, alternative assets include commodities, energy, precious metals, and real estate. There are certainly other alternative assets, but this analysis considers only these four.

In this study, the performance of commodities is represented by the S&P GSCI Commodity Index (prior to February 6, 2007 known as the Goldman Sachs Commodity Index). The performance of energy is represented by the Vanguard Energy Fund, precious metals by the Vanguard Metals and Mining Fund, and real estate by the Fidelity Real Estate Fund. These particular funds were selected because of their long performance histories.

As the term would suggest "alternative" assets are garnishments to a portfolio, rather than a portfolio's core assets. The core assets in this study included large US equity, small US equity, international growth equities, international value equities, US bonds, and US cash. Rather than use the performance of raw indexes, this analysis used the performance of actual mutual funds to represent the core assets. The "core" mutual funds in this study were Vanguard 500 Index, Vanguard Small Cap Index, Vanguard International Growth, Vanguard International Value, Vanguard Total Bond Index, and Vanguard Money Market Prime. All of these funds had performance histories going back to January 1, 1987. The time frame of this study was the 20-year period ending on December 31, 2006.

The primary objective of this study was to identify the effect of adding these four alternative assets to a core equity/bond portfolio in accumulation mode (pre-retirement build-up phase) and during the withdrawal mode (retirement draw-down phase). In this analysis, the impact of adding commodities to a core portfolio was examined separately from adding energy, precious metals, real estate, and a combination of the three to the core portfolio.

Why make these comparisons? First, actionable commodities-based mutual funds and ETFs are a relatively new product despite the fact that annual performance data for the S&P GSCI is available back to 1970. On the other hand, actionable mutual funds that specialize in real estate (i.e. REIT's), precious metals, and energy stocks have long histories. Secondly, mutual funds that attempt to track commodities indexes (such as the S&P GSCI Commodity Index or the Dow Jones AGI Commodity Index) use structured notes and derivatives which creates a far more complicated fund structure. Thus, this study examines the historical impact of adding energy, precious metals, and/or real estate to a portfolio in lieu of commodities. Simply put, is energy, real estate, or precious metals a reasonable surrogate for commodities? All performance data utilized in this study were obtained from Morningstar Principia.

First, some background on the S&P GSCI Commodity Index. As of late May, 2007 the composition of the index was 70.1% energy, 11.35% industrial metals, 2.3% precious metals, 11.39% agriculture products, and 4.87% livestock (see http://www2.goldmansachs.com/gsci/ for current information on the index). As energy comprises the largest single component it was logical to assume that an energy-based mutual fund might serve as an appropriate surrogate for the S&P GSCI Commodity Index.

So, why consider real estate or precious metals as an alternative to commodities? The case for including real estate as an "alternative" asset is strong because of impressive historical returns. The case for precious metals (PM) is modest at best. However, as industrial and precious metals are included in the S&P GSCI Commodity Index (albeit a relatively small percentage) PM becomes a candidate as a commodity surrogate. Secondly, PM has a long history (though perhaps not a wide history) of usage as an alternative asset.

A Tale of Three Portfolios

In this analysis, the value of adding alternative assets to a core portfolio was studied in three different core portfolios: conservative, moderate, and aggressive. As shown in "The Portfolio Lineup" the conservative portfolio has a 40% equity/60% fixed income allocation, the moderate portfolio a 60%/40% mix, and the aggressive portfolio has a 90%/10% allocation.

When alternatives were added (either as commodities, energy, real estate, precious metals, or a mixture of the latter three) the equity allocations in each portfolio were reduced accordingly to create the needed space for the alternative asset. The alternative asset allocation in the conservative portfolio was 10%, 20% in the moderate portfolio, and 30% in the aggressive portfolio.

We look first at the accumulation portfolios (see "Pre-Retirement Accumulation Phase"). In the aggressive 90/10 portfolio, adding an alternative asset of any kind produced a higher return than in the core portfolio. Energy had the greatest impact over this particular 20-year period ($104,786 vs. $78,127). However, adding commodities improved the worst one-month percentage loss more dramatically than any of the other alternative assets. Adding a composite mix of energy, real estate, and precious metals to the core portfolio reduced the worst one-year calendar loss to -8.8% compared to -16.6% in the core portfolio without alternative assets.

In the accumulation phase moderate 60/40 portfolio, adding alternative assets was beneficial in terms of return and reduction in account value losses when measured monthly or annually. The exception was precious metals. While PM did generate a higher return than the core portfolio, the worst monthly loss was higher. Interestingly, if measuring maximum loss over one-year periods (i.e., calendar years) adding precious metals led to the best outcome (-3.6% versus -9.3% in the core portfolio).

Now we turn our attention to the more fragile accounts-retirement portfolios that are sustaining periodic withdrawals (see "Retirement Draw-down Phase"). In this analysis, the accounts were simulated by depositing $10,000 and then withdrawing $75 each month over the 20-year analysis period.

The conservative 40/60 core portfolio without alternative assets ended the 20-year period with a slight loss of principal and had a worst case one-month account value loss of -8.6%. Adding a 10% allocation to commodities modestly improved the ending account value, but significantly improved the worst one-month loss to -6.1% and the average monthly loss to -1.2% from -1.6%. Adding energy to the portfolio had the greatest impact on performance, but did not improve the downside resistance (as measured by worst one month % loss, percentage of months with a loss, or average monthly % loss).

The same patterns were observed in the moderate 60/40 drawdown portfolio with a 20% allocation to alternative assets. Adding commodities led to the largest reduction in account value losses, while adding energy had the greatest impact on performance.

Adding commodities to a portfolio exerts a noticeable impact-most noticeably in the improvement of worst one-month losses in both accumulation and drawdown portfolios. However, over one-year periods, the non-commodity assets provided better loss protection in an accumulation portfolio. In drawdown portfolios, adding commodities provided more downside protection than the other three alternative assets. The reason is low correlation between commodities and the core portfolio assets.

The average 12-month rolling correlation between the monthly returns of commodities and the aggregate monthly returns of the moderate core portfolio assets was 0.05 over this 20-year period. By comparison, the correlation of the moderate core portfolio to energy was 0.61, to real estate 0.51, to precious metals 0.33, and to the alternative asset blend (50% energy, 30% real estate, 20% precious metals) was 0.67. Simply put, commodities zig when the core portfolio assets zag. And while commodities may not produce the highest dollar gain compared to the other "alternative" assets, the low correlation of commodities tends to provide attractive downside protection.

Ironically, the case for adding commodities is stronger in a retirement draw-down portfolio. Minimizing losses is one of the primary goals in a retirement portfolio, and commodities produced the best results in that category. (Actually, low correlation assets provide downside resistance. In this study, commodities happens to be the lowest correlation asset). In accumulation portfolios, the other three alternative assets presented a stronger case for inclusion in a portfolio that did commodities.

Energy and/or real estate present themselves as viable alternatives to commodities. Both provided return enhancement in excess of commodities in the accumulation portfolios (when downside protection is not as vital) as well as loss protection during the drawdown phase. Admittedly, the loss protection was inferior to that provided by commodities. However, given the demonstrated logistical challenges in the creation and delivery of commodity-based mutual funds and ETFs, energy and/or real estate are attractive "alternatives".

The Portfolio Lineup

Asset Class	Conservative 40/60 Portfolio Allocations (%)		Moderate 60/40 Portfolio Allocations (%)		Aggressive 90/10 Portfolio Allocations (%)
	Core Portfolio	Core Portfolio with Alternative Assets	Core Portfolio	Core Portfolio with Alternative Assets	Core Portfolio	Core Portfolio with Alternative Assets
Large US Equity	20	15	25	15	35	25
Small US Equity	10	7	15	10	25	15
Intl Growth Equity	5	4	10	7	15	10
Intl Value Equity	5	4	10	8	15	10
Alternative Asset	0	10	0	20	0	30
Bonds	45	45	30	30	10	10
Cash	15	15	10	10	0	0

Pre-Retirement Accumulation Phase (1/1/1987 - 12/31/2006)

Portfolio	20-Yr Annualized Return	Growth of $10,000 lump sum	Worst One-Month % Loss in Account Value	Worst One-Year % Loss in Account Value
Aggressive 90/10 Portfolio
Core Portfolio	10.83	78,127	-21.3	-16.6
Core + Commodities	11.03	81,014	-13.6	-15.1
Core + Energy	12.46	104,786	-20.9	-11.2
Core + Real Estate	11.69	91,335	-17.9	-9.4
Core + Precious Metals	11.28	84,848	-22.5	-10.5
Core + Alternative Asset Blend (50% Energy, 30% REIT, 20% PM)	12.08	97,880	-20.3	-8.8
Moderate 60/40 Portfolio
Core Portfolio	9.54	61,921	-13.1	-9.3
Core + Commodities	9.56	62,138	-8.1	-8.0
Core + Energy	10.59	74,917	-13.0	-5.2
Core + Real Estate	10.03	67,593	-11.0	-4.1
Core + Precious Metals	9.86	65,621	-14.1	-4.6
Core + Alternative Asset Blend (50% Energy, 30% REIT, 20% PM)	10.31	71,222	-12.6	-3.6

Retirement Draw-down Phase (1/1/1987 - 12/31/2006)

Portfolio	Ending Account Value ($10,000 Initial Sum)	Worst One Month % Loss	Percentage of Months with a Loss	Average Monthly % Loss
Conservative 40/60 Portfolio
Core Portfolio	9,584	-8.6	43.5	-1.6
Core + Commodities	10,245	-6.1	45.2	-1.2
Core + Energy	11,272	-8.5	42.3	-1.5
Core + Real Estate	9,055	-7.5	46.0	-1.3
Core + Precious Metals	9,029	-9.0	45.6	-1.5
Core + Alternative Asset Blend (50% Energy, 30% REIT, 20% PM)	10,191	-8.3	43.1	-1.5
Moderate 60/40 Portfolio
Core Portfolio	13,927	-13.8	39.7	-2.2
Core + Commodities	15,767	-8.8	39.8	-1.6
Core + Energy	18,671	-13.7	41.8	-2.0
Core + Real Estate	13,292	-11.7	39.3	-1.9
Core + Precious Metals	12,575	-14.7	44.4	-2.2
Core + Alternative Asset Blend (50% Energy, 30% REIT, 20% PM)	16,009	-13.3	38.1	-2.1

____________________________________________________________________________________
Craig L. Israelsen, Ph.D. is an associate professor at Brigham Young University. He teaches family finance in the Department of Home and Family Living. His research interests include mutual fund analysis. He writes monthly for Financial Planning magazine. Learn more about Craig Israelsen