This article explains a semi-monthly tactical strategy that combines three of my previous strategies with slight modifications into a unified strategy for larger ($100K - $1000K) retirement accounts. Up to four ETFs are traded at the beginning of each period. Backtesting to the beginning of 2008 shows this strategy produces low annual volatility (< 9%), low maximum drawdown (< 9%), positive annual growth every year, and high compounded annual growth rates (CAGR ~ 20%). The Sharpe Ratio is 1.85 and 70% of the periods have positive growth.
I call this strategy the Unified Tactical Strategy, or UTS. It combines active risk mitigation to avoid large drawdowns while, at the same time, providing significant annual growth and no negative-growth years. UTS is a much better way to invest retirement savings than the more common "buy and hold" strategy used by many investors and advocated by many financial advisors. No one wants to live through another 50% drop in retirement savings, right?
There are three sub-strategies that are incorporated into the UTS. The first two sub-strategies consist mainly of equities with a few other assets such as real estate and gold commodities. The best asset(s) are selected each period based on their relative performance to each other. Provisions are made to shift to the best bond ETF when the top-ranked assets (based on relative strength ranking) do not pass a moving average filter. The third part of UTS is a bond-only sub-strategy consisting of relatively low-volatility bonds. I will explain each of the sub-strategies below, followed by the backtested results of the complete UTS.
The first sub-strategy is the AllAssetsExceptBonds Strategy. Details of this strategy have been reported previously. This sub-strategy ranks and selects from a diversified list of ETF assets including U.S. equities, foreign equities, emerging market equities, real estate and gold commodities. Backtested to 2003, the AllAssetsExceptBonds Strategy has a CAGR of 20.0%, a volatility of 15.0%, a maximum drawdown of 17.3%, and a Sharpe Ratio of 1.12. These numbers are slightly different than those reported in the previous article due to slight variations in the model. The best two assets are selected for each semi-monthly period based on relative growth and volatility. A four-month moving average filter is employed to reduce risk.
The second sub-strategy is the SSSEquity2 Strategy with modifications. Details of the SSSEquity2 Strategy can be found here. I call this new strategy (that incorporates modifications to the SSSEquity2 Strategy) the SSSEquity4 Strategy since it has four fast-growing equity ETFs in its universe: 1) Guggenheim Spin-off ETF (CSD), 2) First Trust IPO Index ETF (FPX), 3) Global X Top Guru Holdings (GURU), and 4) PowerShares Intellidex U.S. Biotech-Genome ETF (PBE). In contrast, the SSSEquity2 Strategy only has three equities: CSD, FPX and GURU.
The relative strength parameters are also different between SSSEquity2 and SSSEquity4. The relative strength categories for SSSEquity4 are six months growth, 20 days growth, and 20-day volatility. The rankings of these three categories are weighted 30%, 35%, and 35% respectively and added together to arrive at an overall ranking of the four equities. The top-ranked equity has to pass a four-month moving average filter before it is selected in a period, otherwise iShares Barclays 7-10 Year Treasury ETF (IEF) is selected for risk aversion. Only one ETF is selected from this sub-strategy.
The backtested results of SSSEquity4 from 2008-2014 using ETFreplay are shown below. The timeframe of the testing is starts in 2008 because of the inception dates of the ETFs. The strategy is compared to the S&P 500 Index ETF (SPY), a benchmark for equities. The CAGR is 27.6%, the volatility is 17.1%, the maximum drawdown is 18.0%, and the Sharpe Ratio is 1.05. The SSS4Equity4 Strategy has positive growth every year, and beats SPY every year, usually by a wide margin.
The third sub-strategy is a bond-only strategy called the LowVolatilityBond, LVB, Strategy. Its ETF universe is composed of: 1) PowerShares S&P-LSTA Senior Loan ETF (BKLN), 2) IEF, 3) AdvisorShares Peritus High-Yield Bond (HYLD), and 4) Barclays Low Duration Treasury ETF (SHY). The ETFs were chosen to counter problematic issues that might occur in the current bond environment. Long duration bond ETFs such as iShares Barclays Long-Term Treasury ETF (TLT), Vanguard Long-Term Government + Credit ETF (BLV), and Vanguard Extended Duration Treasury ETF (EDV) were not included because of their excessive volatility.
Only one ETF is selected each period. 3-month and 10-day returns are used to rank the ETFs, and weightings of 40% and 60% are applied to the respective rankings to arrive at a top-ranked ETF in a given period. The top-ranked ETF has to pass a 2-month moving average filter in order to be selected. Otherwise, SHY is selected.
In order to backtest the LVB Strategy, I had to use a proxy for HYLD, and the proxy I used was Vanguard High-Yield Bond Mutual Fund (VWEHX). There was no proxy available for BKLN, so it was only included in the strategy after its inception date (2011). The backtested ETFreplay results for the LVB Strategy with VWEHX as a proxy for HYLD are shown below for the years 2004-2014. The benchmark is iShares Core Total U.S. Bond ETF (AGG). It can be seen that the CAGR was 12.2%, the volatility was 5.0%, the maximum drawdown was 4.6%, the Sharpe Ratio was 2.05, and 73% of the periods had positive growth. The annual growth beat AGG every year, and in some years did so quite convincingly.
I divided the three sub-strategies in UTS in the following proportions: 30%, 30%, and 40%, respectively. The cash ETF is SHY. The backtested results from 2008-2014 for the complete UTS is shown below. The benchmark is Vanguard 60-40 Balanced Mutual Fund (VBINX), one of the best mutual funds for a 60/40 mix of equities and bonds. The results were presented at the beginning of this article. UTS soundly beats VBINX every year, and had a maximum drawdown of 8.6% compared to VBINX's maximum drawdown of 34.4%. $100,000 invested in the UTS would be worth $303,000 today, while the same amount would be worth only $146,000 if it was invested in VBINX.
One of the ways to increase growth in a tactical strategy is to have a variable cash filter rather than a constant cash filter for the equity sub-strategies. Usually, money is placed in a safe cash ETF like SHY when the top-ranked asset(s) cannot pass a filter such as their 4-month moving average. But it can be shown that improved returns are possible if the cash asset is a variable that depends on the current market conditions rather than a constant ETF like SHY. And SHY may not be the best choice for cash if interest rates increase.
Generally, in the past, more growth in a portfolio can be captured when the market goes bearish if money is transferred into mid- or long-term treasuries (like IEF or TLT) rather than short-term treasuries (like SHY). But there are times when it would be better to put the "risk off" money into other types of bond ETFs. This is especially true in the current market environment with the tapering by the Federal Reserve and the possibility of higher long-term rates (and eventually higher interest rates).
My recommendation is to use a variable cash ETF, letting it be determined by the best bond selected by the LVB Strategy. Although ETFreplay does not have the built-in capability to analyze this methodology, I have looked at the detailed results and significant growth improvements are possible if this methodology is implemented. I will use this approach, i.e. a variable cash ETF, when I make my recommendations for UTS. I would expect CAGR's for UTS to be more like 25% instead of 20% when this methodology is employed.
The current recommendations (as of February 26, 2014) for UTS are: GURU (15%), HYLD (40%), S&P MidCap 400 SPDRs ETF (MDY) (15%), and PBE (30%). Recommendations for each semi-monthly period can be found on my Instablog the evening before the first day of a new period. The next trade day is Monday, March 3, 2014. Recommendations will come out the preceding week-end.