- This article is a sequel to a previous SA article on simple equity-treasury hedge strategies that rely on the negative correlation between treasuries and equities in general.
- In this article, simple tactical modifications have been made to the original strategy to protect it against bear treasury and bear equity markets when they occur at the same time.
- The bond part of the strategy (TMF) and the equity part of the strategy (SPY or SSO) switch to cash (SHY) in bear markets.
- For the TMF/SPY strategy, the optimum split is 25%/75%. Backtesting to 2009 shows a CAGR of 18.0% and a maximum drawdown of 8.4%.
- For the TMF/SSO strategy, the optimum split is 35%/65%. Backtesting shows a CAGR of 26.7% and a maximum drawdown of 12.7%.
Please use this information at your own risk.
Since my last article on simple hedge strategies involving a single equity ETF and a single long-term treasury ETF (see here), I have looked at ways to improve the strategy. The original strategy should do well during either a bull equity market or a bull treasury market, or if both markets are bullish. However, some investors mentioned the need to have protection and reduce drawdown when the equity market and the treasury market are bearish at the same time. This protection can be accomplished by adding simple tactical measures to the basic hedge strategy. This article will present the improvements I have incorporated into the original hedge strategy by adding simple tactical methods.
In my previous article, I introduced simple buy & hold hedge strategies in which a single equity, such as SPDR S&P 500 Trust ETF (NYSEARCA:SPY), and a single long-term treasury are held. Re-balancing occurs periodically to maintain the optimum split between equity and treasury. The treasury ETF that I like the best is to short Direxion Daily 30-Year Treasury Bear 3x Shares ETF (NYSEARCA:TMV). The next best treasury hedge is to long Direxion Daily 30-Year Treasury Bull 3x Shares ETF (NYSEARCA:TMF). For relative strength backtesting using ETFreplay, I can only use the long TMF method. So I will stick with the long TMF ETF for the bond part of the strategy in this article.
The equity strategy I like best is to short ProShares UltraShort S&P 500 ETF (NYSEARCA:SDS). But again, I cannot use a short SDS position in relative strength backtesting in ETFreplay. So, I will go with a ProShares Ultra S&P 500 ETF (NYSEARCA:SSO) position for the equity part of the strategy. To reduce volatility and maximum drawdown (at the expense of growth), I will also look at a long SPY position for the equity position.
I looked at ways for protection against bear markets. The easiest and simplest way to accomplish this is to switch the TMF position to cash, such as iShares 1-3 Year Treasury Bond ETF (NYSEARCA:SHY), when the treasury market goes down. And likewise, the equity position is switched to cash when the equity market goes bearish. In this way, we should avoid large drawdowns in any type of market.
Tactical Bond Sub-strategy
For the bond part of the strategy that uses TMF, I added SHY to the universe, and I switch between TMF and SHY. The top-ranked ETF is picked each semi-monthly period, based on the ETF with the best 3-month return. In addition, a 7-month moving average filter is employed; if TMF is the top-ranked ETF, it must pass through the 3-month filter for it to be selected. Otherwise, SHY is selected.
The ETFreplay results of this bond sub-strategy are presented below. The time frame (April 2009-present) is limited by the start date of TMF. The sub-strategy is compared to TMF by itself in the figure. It can be seen that the CAGR is 21.7%, and the sub-strategy selects TMF in all of the significant upturns and SHY in the downturns. The only time when the strategy is deficient is the April-May 2013 time frame, but the strategy quickly reverts to SHY to avoid excessive drawdown.
Tactical Equity Sub-strategy
For the equity part of the strategy, I use a simple 5-month moving average as a filter. If the equity fails to pass the filter, then the money goes to cash. The ETFreplay results of this equity sub-strategy are shown in the figure below for the 2003-present time frame. The equity in this case is SPY. The CAGR is 10.4%, the volatility is 11.2%, the maximum drawdown is -13.3%, and the Sharpe Number is 0.72.
TMF/SPY Tactical Hedge Strategy
When I combine the bond and equity sub-strategies, I find that the optimum split (based on the Sharpe Number) is 25%/75% TMF/SPY. The ETFreplay results for this combination from 2009 to the present are shown below. Updates are semi-monthly. The CAGR is 18.0%, the volatility is 11.4%, the maximum drawdown is 8.4%, and the Sharpe Number is 1.42. For comparison, buying and holding SPY produces a CAGR of 17.1%, a volatility of 18.4%, a maximum drawdown of 27.1%, and a Sharpe Number of 0.89. Thus, this tactical hedge strategy increases growth and reduces maximum drawdown compared to SPY in the 2009-present time frame. And protection is built in to go to cash when bear markets are present.
TMF/SSO Tactical Hedge Strategy
When I combine TMF with SSO, the optimum split is 35%/65% TMF/SSO. The ETFreplay results presented below are: CAGR=26.7%, volatility=18.1, maximum drawdown=12.7%, and Sharpe Number=1.35. The use of SSO instead of SPY increases the growth of the strategy at the expense of slightly higher volatility and drawdown. Investors must decide for themselves if the risk is worth the reward in going to SSO rather than SPY.
I believe even better returns (at the same drawdown) will occur if a short position of SDS is used as the equity position, instead of a long SSO position. But I do not have the tool to test this idea out.
TLT/SPY Tactical Hedge Strategy
To show how this tactical hedge strategy might perform in large bear markets such as 2008, I have replaced TMF with iShares 20+ Year Treasury Bond ETF (NYSEARCA:TLT). The ETFreplay calculations for the bond portion of the strategy using TLT from 2003-present are shown in the figure below. In general, this simple tactical bond method that switches between TLT and SHY puts this sub-strategy in the proper ETF as the treasury market moves between bear and bull status. The growth of a TLT strategy will, of course, be inferior to the growth of a TMF strategy. But I'm doing this exercise to show how these hedge strategies would have performed in a large bear equity market like 2008.
When TLT is used as the hedge treasury (instead of TMF) and it is combined with the SPY tactical sub-strategy, ETFreplay can be run all the way back to 2003. The backtest results are shown below. These results include the effects of the tactical measures I have discussed previously in this article. The optimum split for TLT/SPY turns out to be 40%/60%. What is important in this calculation is not the CAGR result (that is quite low), but the drawdown in 2008. It can be seen that the maximum drawdown is -8.6%. This exercise gives some confidence that the TMF/SPY and TMF/SSO strategies should avoid excessive drawdowns in large bear equity markets like what we have seen in the past. Of course, a future bear equity market might be concurrent with a bear treasury market, but the strategy should be able to handle that situation too because of risk-mitigation measures incorporated in the strategy.
To conclude, this article has presented a simple tactical hedge strategy that uses only three ETFs: an equity (SPY or SSO), a long-term treasury, and cash. The success of this tactical hedge strategy is based on the generally negative correlation between treasuries and equities. And even if the correlation becomes positive, e.g. at the times when both the equity and treasury markets are bearish, the simple tactical measures incorporated in the strategy should avoid excessive drawdowns by switching to cash. In times when either the equity market or treasury market is bullish (or both are bullish), this strategy should provide good growth and low drawdown in retirement accounts.
I plan on posting semi-monthly updates for this tactical hedge strategy, THS, on my Seeking Alpha Instablog. Currently, the strategy selects SPY (or SSO) and TMF. The splits are 75% SPY and 25% TMF, or 65% SSO and 35% TMF.
Disclosure: The author is long SSO, TMF, SPY. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it. The author has no business relationship with any company whose stock is mentioned in this article.