This is the fourth article in a series of articles on Tactical Asset Allocation (TAA) strategies for retirement savings. As I have tested and studied various TAA strategies, it has come to my attention that many times relatively minor changes to one parameter in relative strength or volatility, or adding one ETF to a large ETF universe can cause large changes in yearly performance and/or drawdown. I work hard to develop strategies that do not exhibit such tendencies. But it begs the question: Is simple better than complex? Maybe it is better to just stay simple in order to more fully understand the interaction of the ETFs and the performance of the strategy, and reduce sensitivity.
So with this mindset, I developed the Super Simple Saving Strategy, i.e. the SSS Strategy. Will it work under all market conditions? Probably not. But the worst that can happen with the SSS Strategy is that the retirement money goes to cash if the ETFs do not perform well or do not pass their moving average filter. And with short-term returns (10 days) and short-term volatility (10 days) included in relative strength rankings, and with moving average filters in place, the SSS Strategy should provide good risk mitigation. And risk mitigation is always a chief concern among investors, especially after the 2001-2002 and 2008 downturns of the market.
I have divided the SSS Strategy into two parts: an equity sub-strategy (SSSEquity) and a bond sub-strategy (SSSBond). SSSEquity is equivalent to the CSD-Bond Strategy I have talked about in a previous article. There are only three ETFs in its ETF universe: Guggenheim Spin-Off ETF (CSD), ishares Barclays 20+ Year Treasuries ETF (TLT) and Barclays Low Duration (1-3 Years) Treasuries (SHY). CSD is selected as the representative equity ETF because of its superior performance over the past five years. TLT is selected because of its negative correlation to CSD (indeed to almost any equity). SHY is basically the cash part of the portfolio in case CSD and TLT are both performing poorly at the same time. Although past history says CSD and TLT move in opposite directions (i.e. negative correlation), we might be in a different market environment these days with long-term interest rates increasing. That is why SHY is used as a safety net.
Using the ETFreplay software, I rank the ETFs using total return (dividends included) of 3 months and 10 days. The ranks are then weighted: 60% for the 3-month ranking and 40% for the 10-day ranking. Then the final ranking is determined by adding the components. And I use a moving average filter of five months for risk mitigation. Updating is semi-monthly (twice per month) to quickly adapt to sudden market changes. The SSSEquity Strategy selects one ETF for each semi-monthly period.
For 2008-2013, the CAGR is 30.9% for this strategy, with a Sharpe Ratio (return-to-risk ratio) of 1.51. Maximum drawdown is 12.9%. The results are shown in the figure below. In comparison, the benchmark S&P 500 ETF (SPY) has a CAGR of 6.2%, a Sharpe Ratio of 0.29, and a maximum drawdown of 51.9%. The backtesting started in 2008 because that was the first year that CSD data was available for the whole year.
A bond sub-strategy (SSSBond) is the second part of the SSS Strategy. SSSBond uses only 4 ETFs: ishares Barclays 7-10 Year Treasuries ETF (IEF), AdvisorShares Peritus High-Yield Bond 3-4 Years (HYLD), PowerShares S&P-LSTA Senior Loan ETF (BKLN), and SHY. IEF and HYLD are selected because they have relatively high return potential for bonds and because of their negative correlation to each other. BKLN is selected as a safety net against short-term interest rates rising in the future (although there is no indication of short-term rates increasing any time soon). SHY is again the cash ETF. The three ETFs (IEF, HYLD and BKLN) are ranked based on total returns (3-months and 10-days), and volatility (10-days). The weighting factors on the ranks are 35%, 35%, and 30% respectively. The top ranking is determined by adding the weighted rankings. The top ranked ETF is selected each semi-monthly period if it passes a two-month moving average filter. If not, cash is selected.
In addition, in order to backtest to 2004, a proxy mutual fund of the high yield bond class is included in backtesting: Vanguard High-Yield Corporate Bond Mutual Fund (VWEHX). To determine the suitability of VWEHX as a proxy for HYLD, backtested results were compared in the 2011-2013 timeframe since HYLD's inception date was 2011. The results are not shown here, but good agreement was seen.
ETFreplay results for the SSSBond Strategy from 2004-2013 are shown in the figure below. The CAGR is 11.8%, the Sharpe Ratio is 1.96, the overall volatility is 4.9% and maximum drawdown is 7.2%. For comparison, the benchmark ishares Core Total US Bond 4-5 Years ETF (AGG) has a CAGR of 4.4%, a Sharpe Ratio of 0.42, an overall volatility of 5.3% and a maximum drawdown of 12.8%. The total return of SSSBond is positive every year and beats AGG every full year.
(click to enlarge)It is important to note 2013 in particular because March 2013 signaled the end of the long-term bond bull market according to most economists. Most bond funds and ETFs lost money in 2013; the only exception were high-yield bond funds/ETFs. For example IEF lost 5.3% in 2013, and AGG lost 1.8%. But the SSSBond Strategy mainly selected HYLD in 2013, and produced a growth of 8.6%. Because HYLD consists mainly of low duration high yield bonds, it will do well even if long-term interest rates increase. It is also a managed fund and has risk mitigation strategies built-in if short-term rates increase.
The entire SSS Strategy was run with a mix of 60% SSSEquity and 40% SSSBond. The ETFreplay results for 2008-2013 are shown in the figure below. The CAGR is 24.8%, the Sharpe Ratio is 1.81, and the maximum drawdown is only 7.4%. $100,000 invested at the beginning of 2008 would have grown to $379,900 at the present time. In comparison, the benchmark Vanguard 60-40 Balanced Fund (VBINX) had a CAGR of 6.4%, Sharpe Ratio of 0.37, maximum drawdown of 34.3%, and a present total worth of $145,600 from an initial $100,000 investment. The yearly growth of the SSS Strategy is positive every year, with a minimum yearly growth of 15.3% in 2008 and a maximum yearly return of 40.8% in 2009. The current ETF selections of the SSS Strategy are 60% CSD and 40% HYLD.
The selections of the SSS Strategy will be posted at the end of every semi-monthly period on my Instablog. I'm hoping Instablog notifies every follower of the posting. If not, then you will have to look at my Instablog at the end of every semi-monthly period. The first posting will be on the evening of January 15th. Selections of my other strategies from previous articles will likewise be posted.