Please realize you use these strategies at your own risk.
Hedging an equity position with long-term treasuries is a way to increase growth and decrease volatility and drawdown of an equity strategy. The reason this works is because of the negative correlation seen between equities and treasuries, especially during bear equity markets. This negative correlation between treasuries and equities has been the norm in bear equity markets (and corrections) from 2000-present.
I decided to try and develop some simple buy-and-hold hedging strategies that do not use tactical switching between assets (except to rebalance periodically). As reported by Frank Grossmann, the best hedging strategy if you hold an equity position is to short the Direxion Daily 30-Year Treasury Bear 3x Shares ETF (NYSEARCA:TMV). Harry Long has also postulated this sort of hedging strategy (e.g. see here).
There are two issues with using a short TMV position as a hedge. The first issue is that inverse ETFs like TMV cannot be bought in retirement accounts. So that limits its applicability. The alternative is to use the less effective Direxion Daily 30-Year Treasury Bull 3x Shares ETF (NYSEARCA:TMF) as the hedge. I will show the difference between using a short TMV position as a hedge and a long TMF position as a hedge later in this article.
A second issue with using a short TMV position is that TMV has only existed since April, 2009. So although we can optimize a split between a short TMV and an equity in the 2009-present timeframe and present backtesting results, there is still the concern of whether the strategy will be effective during prolonged bear equity markets like 2001-02 and 2008. I will get back to this concern in a moment.
First, I want to show that this basic hedging strategy works well in backtests from 2009-present. I looked at four different strategies, going from the conservative to the aggressive. I first studied a short TMV / long SPDR S&P 500 Index (NYSEARCA:SPY) strategy. Using the ETFreplay software, I ran a series of splits between short TMV and long SPY, and I found that the Sharpe Ratio was optimum with a 20/80 short TMV/long SPY split. The backtested results from 2009-present are presented below. The CAGR is 21.1%, the volatility is 11.3%, the maximum drawdown is 9.0%, and the Sharpe Ratio is 1.57. These are all good numbers, and based on the results from 2009-present, this strategy seems to be a good way to avoid excessive risk and produce good annual growth. For comparison, buying and holding SPY produced a CAGR of 18.7%, a volatility of 16.6%, a maximum drawdown of 18.6%, and a Sharpe Ratio of 1.04. It is significant that volatility and drawdown are significantly reduced when the hedging strategy is employed versus when hedging is not used. And this occurs, interestingly, even though the volatility of TMV is three times the volatility of SPY.
To show the effect of hedging with a long TMF position rather than with a short TMV position, I reran this strategy with a long TMF position. The results (presented below) show a CAGR of only 18.4%, compared to 21.1% when a short TMV position is used. This is a sizable difference and shows why a short TMV position is much better than a long TMF position as a hedge. Grossmann has shown this same effect on his blog.
A second hedging strategy I investigated was a strategy that hedged a short -ProShares Short S&P 500 ETF (NYSEARCA:SH) with a short TMV. This is the same strategy as the first strategy except I have substituted a short inverse equity index ETF for a long position in that same index. The use of a short inverse equity position (instead of a long position) increases the growth of the strategy without any other negative effects. The backtest results of this strategy is shown below. Again, the optimum split is 20/80 as in the first strategy. The CAGR has increased to 22.7% (compared to 21.1% in the first strategy). The volatility is 11.3%, the maximum drawdown is 8.8%, and the Sharpe Ratio is 1.62. So going to a short inverse equity position instead of a long position has bought us about 1.6% in CAGR with little other effect.
A third simple hedging strategy I studied was to use a short position of ProShares UltraShort S&P 500 ETF (NYSEARCA:SDS) hedged with a short position of TMV. Alternatively, I could have used a long position of ProShares Ultra S&P 500 ETF (NYSEARCA:SSO), but the use of the short position of SDS increased the CAGR by 5% without any negative effects to the strategy. So I chose to use a short SDS position instead of a long SSO position.
The backtest results (2009-present) are shown below. The optimum split is 35/65 for short TMV / short SDS. The CAGR is 39.7%, the volatility is 18.6%, the maximum drawdown is 14.4%, and the Sharpe Ratio is 1.62. As expected, the increase in growth came at the expense of volatility and drawdown. Still, the volatility is not much more than SPY by itself, and the drawdown is less than SPY (for 2009-present). For the moderate investor, this strategy may be appealing.
A fourth strategy that I studied was focused on maximizing CAGR. This strategy used a short TMV position with a short position of Direxion Daily S&P 500 Bear 3X Shares ETF (NYSEARCA:SPXS). This is a very aggressive strategy that only the most ambitious investors will want to utilize. The ETFreplay results are shown below. The CAGR is 54.5%, the volatility is 23.9%, the maximum drawdown is 20.1%, and the Sharpe Ratio is 1.62. It should be stated that using a short position in SPXS increased the CAGR by 9% compared to using a long position of Direxion Daily S&P 500 Bull 3X Shares ETF (NYSEARCA:SPXL). The optimum split is 45/55 for short TMV / short SPXS.
But the question still remains, how will these hedged strategies perform in a large bear equity market, something like 2001-02 or 2008? We cannot actually backtest TMV to those years, but we can use a long position in the Vanguard Long-term Treasury Mutual Fund (MUTF:VUSTX) as a poor man's treasury hedge against SPY. VUSTX goes back beyond 2000 so we can backtest with it to 2000 in ETFreplay. In this case, VUSTX is a proxy for iShares 20+ Year Treasury Bond ETF (NYSEARCA:TLT) to enable backtesting to 2000. I will now try to show that another bear equity market scenario will not be overly detrimental to the hedge strategies I am proposing. I will concentrate on the first hedge strategy, the one that uses short TMV / long SPY, and I will replace a short TMV position with a long VUSTX position.
There is one part of the 2009-present timeframe that mimics a bear market in a minimal way: the period from July, 2011 to January, 2012. The hedge strategy needs to come through this market correction period without significant drawdown for us to have some confidence in the effectiveness of the hedge strategy in a bear equity market. In the first figure presented in this article, it can be seen that the strategy has low drawdown in this period.
But if I use a long VUSTX position as a replacement for the short TMV position and I use the same splits as was employed with the short TMV position (i.e. 20/80 long VUSTX/long SPY), it can be seen from the figure below that the July, 2011 - January, 2012 period is not well-captured by the long VUSTX/long SPY strategy as compared to the short TMV/long SPY strategy. For a 20/80 split, VUSTX does not have enough effect to compensate for the downturn in SPY in this time period. A different split is needed.
So I proceeded to backtest the long VUSTX / long SPY strategy to 2000, and it turns out the optimum split is 65/35. This split makes some sense since TMV is leveraged 3X from VUSTX. The ETFreplay results are shown below. With the long VUSTX / long SPY strategy, I can backtest to 2000. The CAGR of this strategy is not very good, but that's not the purpose for showing these results. The purpose is to show that a hedging strategy can work through the bear markets of 2001-02 and 2008. It can be seen that the maximum drawdown is about 16% for this strategy compared to 55.3% for the S&P 500. I think this exercise should give us some confidence that all four of the hedge strategies presented in this article should significantly reduce drawdown in bear equity markets, while giving us significant annual growth.
The only caveat I would throw out with this hedge strategy is my concern about how these strategies will perform if the negative correlation between treasuries and equities is not maintained in a bear equity market. If both the equity and treasury positions lose money at the same time, then a large drawdown can occur. From 2000-present, the negative correlation between treasuries and equities was maintained in bear equity markets (and even in equity market corrections). But with the potential of rising rates as well as the potential of an equity correction and/or bear market, there might be a possibility of both treasuries and equities moving downward at the same time. If this happens, these strategies may not be effective to reduce risk.
So, again, please use these strategies at your own risk.
Disclosure: The author is short TMV, SH, SDS, SPXS. 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.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.