Alternatives: The Underused and Underappreciated Diversification Tool – Part 2

This is Part 2 in a series on the use of Alternative investments as a diversification tool.  This article examines the risk and return characteristics of long-only and Alternative funds.

We base our diversification work on six long only asset classes: Bonds (both U.S. and international), Global Equities, Commodities, and Real Estate (U.S. and international). We believe these investments are a good representation of what financial theory would refer to as the set of all assets available on a global basis.  We aim to activate all six long-only asset classes with one or more Alternative investments.

Below is a list of the long-only mutual funds and ETFs used in our article, followed by a list of the types of Alternative investment categories that were used.

Long-Only Funds and ETFs used in this article:

• iShares Dow Jones U.S. Real Estate ETF (IYR)
• Fidelity International Real Estate (FIREX)
• Barclay’s Aggregate Bond ETF (AGG)
• GSCI Commodity ETF (GSG)
• Vanguard Global Equity (VHGEX)
• T. Rowe Price International Bond (RPIBX)

Alternative Investment Categories based on Lipper Classifications (we will use the classifications averages to proxy for each strategy):

• Long/short
• Long/short equity
• Long/short bond
• Global macro
• Multi-strategy
• Market-neutral
• Arbitrage
• Managed futures
• Commodities
• Quant
• Currency
• Event-driven

The evaluation metrics we use for both Alternatives and long only assets are: return, volatility, downside deviation, both the Sharpe and Sortino ratios, skewness, and excess kurtosis. To ensure diversification is being improved, the metrics for each existing long-only asset class are computed both before and after Alternatives are added. This assures that the addition of Alternatives is judged fairly.

Baseline Metrics: March 2007-September 2012

Table 1 shows the evaluation metrics for buy-and-hold investments, while Tables 2A and 2B show the same metrics for Alternatives:

Table 1: Risk-Return Values–Buy-and-Hold Investments

 NB: InTables 1 through 6B, return values have been annualized as has volatility (standard deviation) and the Sharpe and Sortino ratios.

Table 2a: Risk-Return Values–Alternatives

Table 2b: Risk-Return Values–Alternatives

Comparing metrics, we see a number of the Alternatives having greater returns, lower volatility, less extreme risk, and often (but not always) better risk-adjusted returns. It is only in asymmetric risk (skewness) that the two groups of investments match up.  Most Alternatives, therefore, could be good diversifiers for the long only assets. The Alternatives that do not serve as good diversifiers are: multi-strategy, managed futures, currency and quant.

We verify these statements by measuring what are called the betas for overall market sensitivity, skewness and kurtosis.

Based on our tests, arbitrage performs well because it reduces volatility and skewness in most long-only asset classes. Its kurtosis or extreme risk, however, is the highest of all the assets. We will still construct arbitrage + passive portfolios, despite the very high kurtosis. Market-neutral also performs well. Long/short equity works, as does global macro. Given our beta tests, we form portfolios with the following active components: arbitrage, market-neutral, long/short equity, and global macro.

Below we show the portfolios formed during the 2007 – 2012 period.  Each period has three portfolios formed though only two are shown: one where 5% of the portfolio is in Alternatives, and another with 30%.  We do not show the 15% portfolios as it occupies a middle ground between the 5% and 30% portfolios, and therefore, we believe, adds little new information to our analysis.

Portfolio Results: March 2007-September 2012

Table 3: Risk-Return Values–Buy-and-Hold Investments, March 2007-September 2012

Table 4a: Risk-Return Values–Buy-and-Hold Investments + Arbitrage Alternatives at the 5% Level, March 2007-September 2012

Table 4b: Risk-Return Values–Buy-and-Hold Investments + Arbitrage Alternatives at the 30% Level, March 2007-September 2012 

At the 5% level there is hardly any change in any of the metrics. At the 30% level, return, volatility, downside deviation and risk-adjusted return are better. Skewness tends to be better, while kurtosis is typically worse, especially for commodities and world equity. This better performance at the 30% level is typical of all the Alternatives tested, i.e. arbitrage, global macro, market neutral, and long/short equity funds, and seems to indicate that a higher allocation to Alternatives can improve the performance of a portfolio more than a lower level of investment in Alternatives.

In the last of article in our series, we will show what the effect would have been on including Alternatives during the credit crisis.