Summary

  • Diversification and trend-following often fail to protect investors in sudden, tail-risk events
  • By being constantly hedged to such events, investors can help manage this risk, but remaining hedged requires paying an often significant “insurance premium”
  • Spot CBOE VIX would be a great potential hedge, but is inaccessible; access through futures comes with large negative roll yield
  • A combination strategy of exchange-traded products like VXX and XIV may give us hedges in tail events without the large decay of naked short-term futures exposure.

 

The market has a long history of crises. In the last 30 years, we’ve witnessed at least 6 sudden and significant market sell-offs that are supposed to be “100 year events”:

  1. On Black Monday, October 28, 1987, the Dow Jones Industrial Average fell 22.16%.
  2. When Russia de-valued the ruble and defaulted on its debt in August 1998 the S&P 500 sold off 19% in two short months.
  3. The day Lehman Brothers went bankrupt – October 15, 2008 – the S&P fell 9.8%.
  4. The Flash Crash of May 6, 2010 caused the S&P 500 to suffered a 10.2% intra-day drawdown.
  5. During two weeks in late July/early August 2011, the S&P 500 fell 10.2% on fears of contagion of the European sovereign debt crisis.

These events occur so quickly, and with such velocity, that standard techniques like trend-following are not necessarily able to get on the right side of the trade fast enough. And diversification? You know what they say: the only thing that goes up in a crisis is correlation.

But what do these periods have in common? Volatility spikes. The most popular measure of market volatility is the CBOE VIX, which is derived from option chains on the S&P 500.

From 6/30/1998 to 8/31/1998, the VIX climbed 126%. While the VIX didn’t budge on the date of the Lehman bankruptcy, from 8/29/2008 to peak on 11/20/2008, the VIX climbed 291%. During the Flash Crash, the VIX spiked 20%. And during August 2011, the VIX rallied over 100%.

Unfortunately, “spot” VIX is an inaccessible investment. But futures on the VIX are available for investors. And ETF investors can actually access these futures through exchange-traded note instruments like the iPath S&P 500 VIX Short-Term Futures ETN VXX.

Similar to spot VIX, from 8/29/2008 to 11/20/2008, VXX was up 326% and in 2011, it spiked 69%.

But caveat emptor: there is a significant cost to owning short-term futures on the VIX when volatility isn’t spiking.

The VIX futures curve is normally in contango, meaning that as the contracts move towards expiration, there is a price decay. That means every month, VXX suffers from significant negative roll yield. Placid market environments can wreak havoc on returns: from 12/30/2011 to 6/30/205, VXX returned a nearly unbelievable -96.45%.

VXX Adjusted Close

In a sense, this negative roll yield is the insurance premium we pay for the massive gains we experience when volatility spikes. But few investors are willing to pay that much for insurance. How many home owners have an insurance premium equal to the cost of their homes?

But where VXX suffers from negative roll yield, the VelocityShares Daily Inverse VIX Short-Term ETN XIV profits from it. On the other hand, since XIV is short VIX futures, it gets crushed in volatility spikes. From 8/29/2008 to 11/20/2008, XIV lost 81%. Which is sort of the opposite of what we’re looking for here in a hedge.

But, wait. If VXX was up 326% and XIV was only down 81% … a 50/50 mix would be up 122.5%. And while VXX is decaying, XIV is profiting to offset it. There might be something to this.

Enter the S&P 500 VIX Futures Long/Short Strategy Index Series.

SNP VIX Futures Tail Risk Index TR

By comparison, this series has been able to hold its value in low-volatility environments far better than just naked exposure to short-term VIX futures.

So how does it work? The index series comes in many flavors, but follows a general pattern: blend a mix of 2x levered VIX futures and 1x inverse VIX futures. Set the weights, let them drift, and bring them back in line every 13 weeks.

In this manner, the index seeks to negate some of the negative roll costs associated with VIX futures while simultaneously taking advantage of the asymmetric response the two legs of the strategy will have to spiking volatility. Then, if a spike occurs right before rebalance, the profits get locked in.

Unfortunately, depending on which week you start this process, you can get some very different results. That makes sense because it all depends on when a crisis occurs.

We’ve replicated these indices (though slightly modified – I’ll get to that in a bit) for the 13 potentially different starting weeks. The amount of “hedge” you get is very, very different.

VXX-XIV Sub-Portfolios

And, as we would expect, which of the potential portfolios performs best is totally random. Below we see that sub-portfolio 3 was one of the worst hedges in 2008 but the best in 2011.

VXX-XIV Sub-Portfolios - 2008VXX-XIV Sub-Portfolios - 2011

The good folks at S&P are well aware of timing risk and actually go ahead and implement portfolio tranching.

To manage this index, S&P creates all 13 possible sub-portfolios. Each sub-portfolio is rebalanced back to its target underlying weights independently on a quarterly basis, but rebalance dates for the sub-portfolios are spread evenly throughout a quarter on a weekly basis.

At the end of each quarter, the indexes exposure to each of these sub-portfolios is rebalanced back to 1/13th.

Now, for our purposes, this methodology has a bit of an issue. Namely that while there exists a 2x VIX short-term ETN, TVIX, it is less liquid than VXX and a bit on the expensive side (1.65% versus 0.89%).

So instead of holding 45% in the 2x short-term VIX futures position and 55% in the 1x inverse VIX futures position, we assumed a portfolio that was 90% short-term VIX futures and 55% inverse VIX futures, creating a total exposure of 145%. We then normalized these exposures to get 62% in the 1x short-term VIX futures position and 38% in the 1x inverse position.

Now, I’ll admit it: such a system is a bit complicated to implement. Trying to coordinate and aggregate up the weights of 13 individual sub-systems into just two positions requires a bit of math.

But the results of the effort are promising.

Consider this volatility hedging strategy against other possible hedges: long-dated U.S. Treasuries (TLT) and anti-beta long/short (BTAL):

Aligned Volatility Hedges

Hedge Performance

All three methods seem to afford us some decent crisis protection.

While the volatility hedge offered some of the most significant protection, it has also decayed the most in value since 2011 in more placid environments.

TLT, on the other hand, has not only not decayed, it has actually rallied significantly post 2011. BTAL decayed, but certainly less quickly than the volatility hedge (down ~20% from 12/31/2011 – 6/30/2015).

But each method comes with its own risks and assumptions.

The volatility hedge assumes that volatility will increase in a crisis or tail event. So far that has proved to be true in most significant sell-offs. But sometimes volatility can retreat from peak levels while the market continues to sell off. See Q1 2009 as an example.

But the biggest risk to this sort of strategy is decline with no volatility. Perhaps a prolonged one, like Japan. But that is, ideally, where trend-following risk management kicks in.

In using TLT, we assume that U.S. Treasuries will remain the asset of safety around the world. While holding this position only for equity hedging can be beneficial since it might actually have positive expected returns in non-crisis periods as well, it is also incredibly sensitive to interest rate changes. Periods of high correlation between stocks and bonds (e.g. the 2013 taper tantrum) or rising rates might be problematic for this hedge.

Finally, with BTAL we assume that low-beta stocks will outperform high-beta stocks in market crises. Based on anti-beta research, we may also be able to expect healthy absolute returns over the long run, but so far those have not materialized.

The reality is, finding a “perfect hedge” is near impossible. The ideal hedge would return hundreds, if not thousands of percent in a crisis, allowing us to dedicate a small amount of capital to it. An ideal hedge would have little to no decay during calm periods. An ideal hedge would offer protection in all crises, no matter the type. But that is a holy grail hedge.

While the VIX-based hedging strategy presented herein may not be that holy grail hedge, it is certainly another strategy worth considering for those concerned about sudden and violent market shifts.

 

If you want to learn more about how we employ this hedging method in our Dynamic Alternatives strategy, reach out at info@thinknewfound.com.

Corey is co-founder and Chief Investment Officer of Newfound Research, a quantitative asset manager offering a suite of separately managed accounts and mutual funds. At Newfound, Corey is responsible for portfolio management, investment research, strategy development, and communication of the firm's views to clients.

Prior to offering asset management services, Newfound licensed research from the quantitative investment models developed by Corey. At peak, this research helped steer the tactical allocation decisions for upwards of $10bn.

Corey is a frequent speaker on industry panels and contributes to ETF.com, ETF Trends, and Forbes.com’s Great Speculations blog. He was named a 2014 ETF All Star by ETF.com.

Corey holds a Master of Science in Computational Finance from Carnegie Mellon University and a Bachelor of Science in Computer Science, cum laude, from Cornell University.

You can connect with Corey on LinkedIn or Twitter.

Or schedule a time to connect.