NEW YORK (Scott's Investments) -- I had a thought tonight while mowing lawn - can individual investors add portfolio alpha by allocating a small percentage of their portfolio allocation to a strategy which rotates into leveraged ETFs? In other words, can we decrease portfolio volatility and increase portfolio returns by rotating a small percentage of our portfolio in and out of leveraged ETFs? I performed a similar test earlier this year but it involved a 50% allocation to leveraged ETFs, which is way too high for most risk-adverse investors. The full results and methodology of that test are available.
First, some important caveats. The results below are over a very limited time period, primarily due to the relatively short trading history of leveraged ETFs. Thus, be careful not to project historical returns into the future. This is an important point that is too often overlooked.
Second, leveraged ETFs carry additional risk. They are intended as short-term trading vehicles and I would recommend reading the prospectus for the two ETFs discussed below (SDS and SSO) at Proshares website. Compounding -- the likelihood of the funds' returns over periods longer than one day to differ from the target returns -- is a critical concept to understand before investing in leveraged ETFs. Finally, commissions and taxes are not included in the results and will impact returns.
Now, on to the fun stuff. The results of these backtests are courtesy of ETFReplay.com (check it out). I started with a basic buy-and-hold portfolio of five ETFs that might represent a typical investor portfolio. The portfolio consists of the following ETFs and I included the month/year trade data was available on each ETF:
- iShares Barclays Aggregate Bond (AGG) - Get Report (Sept. 2003)
- PowerShares DB Commodity Index (DBC) - Get Report (Feb 2006)
- iShares MSCI EAFE Index (EFA) - Get Report (Aug 2001)
- iShares MSCI U.S. REIT (VNQ) - Get Report (Oct 2004)
- Vanguard MSCI Total U.S. Stock Market (VTI) - Get Report (May 2001)
If an investor had bought and held this portfolio in equal weight, the returns since 2006 are below. Note that for the beginning of 2006 DBC was not held and the portfolio was re-balanced February 15, 2006 to include DBC in the returns:
Total Return: 29.1% vs 4.6% for SPY
Volatility: 20.7% vs 24.4% for SPY
Compound Annual Growth Return: 4.6% vs .8% for SPY
Sharpe Ratio: .13
Correlation to SPY: .92
Strategy drawdown: -45% vs. -50.9% for SPY
A diverse buy-and-hold portfolio still suffered significant portfolio drawdowns in 2008. It has taken two to three years to gain back what was lost in 2008. The portfolio still outperformed
(SPY) on returns and volatility, but remained highly correlated to SPY and I suspect that the individual investor may not be satisfied with an equity curve resembling a thrilling roller coaster.
Next, I took the same buy-and-hold strategy and allocated 90% of my hypothetical total portfolio to it. I then used 10% of my hypothetical portfolio and allocated it to a relative strength strategy that rotates between
Barclays Low Duration Treasury
-2x ProShares Leveraged Short S&P 500
+2x ProShares Leveraged Long S&P 500
(SS0). The strategy (which I will refer to as "leveraged rotations strategy") involves buying the single ETF among these three which has the highest ranking on a semi-month basis. The ranking is determined by the three-month returns, 20-day returns, and 20-day volatility of each ETF. Three-month returns are given a 40% weighting, 20-day returns a 30% weighting, and 20-day volatility a 30% weighting.
First, for information purposes, the results of the leveraged rotation strategy on its own:
Total Return: 73.4% vs 4.6% for SPY
Volatility: 40.5% vs 24.4% for SPY
Compound Annual Growth Return: 10.3% vs .8% for SPY
Sharpe Ratio: .33
Correlation to SPY: -.39
Strategy drawdown: -31.6% vs. -50.9% for SPY
On its own this strategy benefited from the singular down move in 2008/early 2009 and the singular up move since March 2009. However, with volatility of 40.5% this is not a "sleep well at night" strategy. Please note that the SSO and SDS did not begin trading until October 2006, so for most of 2006 the strategy was in cash.
What if we combine this strategy with a buy and hold strategy? Can we increase returns and lower volatility? Below are the results if we allocate 90% of our portfolio to the buy-and-hold strategy and 10% to the leveraged rotation strategy. The start date for the test was 2006:
Total Return: 38.8% vs 4.6% for SPY
Volatility: 16.9% vs 24.4% for SPY
Compound Annual Growth Return: 6% vs .8% for SPY
Sharpe Ratio: .21
Correlation to SPY: .88
Strategy drawdown: -38.5% vs. -50.9% for SPY
We see a slight increase in returns and lower volatility. However, we still see a significant portfolio drawdown but less than buy-and-hold by itself. Given that the strategy incurs several (92) trades (and potentially commissions) over 6-7 years, the added returns could be negligible, depending on the portfolio size.
If we increase exposure to the leveraged rotation strategy to 20% while holding 80% in a buy-and-hold strategy, total returns increase to 46.9% and volatility decreases to 14.7% when compared to a 90/10 allocation:
The added benefit of rotating into leveraged ETFs shows very limited promise but not enough for me to rush out and implement this strategy with my own money. In addition, the short trading history of these vehicles means it is too early to offer any conclusions with confidence. In the coming days I will analyze and publish results of portfolios combining leveraged/short ETFs in combination with other portfolio strategies.
Follow Stockpickr on
and become a fan on
-- Written by Scott Rothbort in Millburn, N.J.
At the time of publication, author had no positions in stocks mentioned.
This article was written by a member of the Stockpickr community. Stock recommendations and comments presented on Stockpickr.com are solely those of the members quoted. They do not represent the opinions of Stockpickr.com on whether to buy, sell or hold shares of a particular stock. Members should be cautious about any and all stock recommendations and should consider the source of any advice on stock selection. Various factors, including personal or corporate ownership, may influence or factor into a member's stock analysis or opinion. All members are advised to conduct their own independent research into individual stocks before making a purchase decision. In addition, investors are advised that past stock performance or portfolio performance is no guarantee of future price appreciation or performance. Furthermore, Stockpickr.com does not guarantee the accuracy or completeness of information on the site, nor does Stockpickr.com assume any liability for any loss that may result from reliance by any person upon any such information or recommendations. Such information recommendations are for general information only.