Shorting Volatility Spikes

Mark Sebastian sent along a piece Russell Rhoads did for the most current (September) issue of Expiring Monthly Magazine. If volatility is something you would like to know more about, we highly recommend looking at that issue (link at bottom of this article). This issue has several great articles on volatility. Thank you Mark for sharing this article with our community.

———

At the beginning of August the VIX ran from 32.00 to 48.00 in a single day. As volatility tend to revert to the mean, many traders that follow and take positions based on the VIX will start to think about a way to benefit from the VIX dropping after such a spike. Keeping this in mind, I did a little research on what may have been the best method of getting short volatility after this recent spike. Below is a VIX chart covering July and August 2011 that clearly shows the price spike I am talking about. 

Along with the VIX index becoming a mainstream topic among market professionals, there has been a dramatic increase in the number of methods available to take a position on the future direction of volatility. Looking back at the price action in the VIX after the run to 48.00, within a week the index was back below 32.00 at 31.87. For comparison sake I took a look at some of more liquid products that current derive their value from the VIX index.

 

Ticker

Instrument

8/8/2011

8/15/2011

% Change

VIX

VIX Index

48.00

31.87

33.60%

VXQ11

August VIX Future

36.55

32.10

12.18%

VXU11

September VIX Future

30.20

26.70

11.59%

VXX

iPath S&P 500 VIX Short Term Futures ETN

34.78

32.18

7.48%

VXZ

iPath S&P 500 VIX Mid-Term Futures ETN

58.12

54.47

6.28%

XXV*

iPath Inverse S&P 500 VIX Short Term Futures ETN

33.27

33.75

1.44%

XIV*

VelocityShares Daily Inverse VIX ST ETN

10.00

10.10

1.00%

 

The table above shows the five day performance for a variety of direct methods for shorting volatility after the quick run to 48.00. Just below the VIX index I show the performance for the first two month VIX futures contracts over the five trading. Although the futures have moved lower, the return is calculated as a positive as the assumption is we took a short position in these contracts. The returns of approximately 12% for both instruments are pretty impressive when view individually, but not in comparison to the 33% drop in the VIX Index. This is typical of volatility spikes where the index moves to a premium relative to the futures contract.

Four exchange traded notes (ETN’s) are listed, two that offer long exposure and two that offer short exposure to the VIX. The VXX focuses on short term volatility taking a balanced position in the front two month futures contracts. The VXZ focuses on volatility a little farther out representing positions in the fourth through seventh month VIX futures contracts. A short position in either of these ETN’s would have a pretty paltry return when compared to the returns from the index or the future contract.

What was really surprising was the return generated from the two inverse ETN’s on the list. As assumption is made that a long position would be taken in either the XXV or XIV based on an outlook for a lower VIX over the near term. Using either of these would actually have returned about 1% over the five day period.

After comparing these methods of trading the VIX, shorting the front month future worked the best with a return of a little over 12%.  Remember that VIX index options trading closely with the VIX futures pricing, so strategies using VIX options would likely have yielded much better returns.

To see if this sort of price action around a VIX spike was common I took a look at what happened the last time we had a quick run in the VIX to the upside. Scanning a chart, there appeared to be a similar instance of a VIX run in March of 2011. A two month chart showing this move appears below. 

The interesting part of this move is that the VIX dropped 34.8% over the five day period following a run up on March 16th. This price drop is pretty much in line with the drop of 33.6% that occurred in the previous example. This is basically just a circumstantial occurrence, but works out nicely to see how our trading vehicles performed over the same period.

 

Ticker

Instrument

3/16/2011

3/23/2011

% Change

VIX

VIX Index

29.40

19.17

34.80%

VXJ11

April VIX Futures

24.50

20.50

16.33%

VXK11

May VIX Futures

25.15

21.60

14.12%

VXX

iPath S&P 500 VIX Short Term Futures ETN

37.63

31.26

16.93%

VXZ

iPath S&P 500 VIX Mid-Term Futures ETN

63.30

56.15

11.30%

XXV*

iPath Inverse S&P 500 VIX Short Term Futures ETN

32.85

34.06

3.68%

XIV*

VelocityShares Daily Inverse VIX ST ETN

11.04

13.20

19.57%

 

First, the two front month futures contracts were down less than the VIX index. A short position in the front month April contract would have returned 16.33%. Shorting the May VIX contract would have resulted in a gain of 14.12%. As in the first example, the futures did not run up as much as the index which again is typical when the VIX moves up in a short period of time.

The results from shorting the two most liquid long ETN’s were a little bit better than in the previous example. Shorting the VXX would have returned 16.93% and a five trading day short position in the VXZ returned 11.30%. Again neither matched the spot index drop, but in this case both compared well versus shorting the futures.

The poor performance was repeated by one of the two inverse VIX related ETN’s. A long position in the XXV would have returned 3.68%. In this instance, the XIV actually performed exceptionally well, gaining 19.57%. This performance makes the XIV the best performing method of trading the move higher in the VIX in March.

Let’s go back one more price run and drop in the VIX to see what seems to work. So far consistency appears to be on the side of the futures contracts. So we can give this one more look by checking out catching the drop in the VIX that occurs right after the market in May 2010. For those with short memories, May 2010 will be remembered by market participants as the occurrence of the ‘flash crash’. The causes of this event are still debated to this day.

The actual date of the ‘flash crash’ was May 6th, but the closing high in the VIX around that price action occurred on May 20th. On the 20th the VIX peaked at 45.79 and then over the next five days it lost around 35% dropping down to 29.68. Again, just by coincidence, the drop is very close in level on a percentage basis as the previous two examples. The comparison table does not include the two VIX related inverse ETN’s as they had not been issued in May 2010.

 

Ticker

Instrument

5/20/2010

5/27/2010

% Change

VIX

VIX Index

45.79

29.68

35.18%

VXM10

June VIX Futures

35.95

29.65

17.52%

VXN10

July VIX Futures

35.70

30.95

13.31%

VXX

iPath S&P 500 VIX Short Term Futures ETN

34.07

28.33

16.85%

VXZ

iPath S&P 500 VIX Mid-Term Futures ETN

94.85

88.88

6.29%

 

The futures contracts have similar performance as in the previous two examples. The front month, June contract moves down just over 17.5% and the second month July contract was down 13.31%. The consistency for both futures is pretty interesting, since theoretically there is not truly a financial relationship. This relationship is non-existent due to the lack of an arbitrage relationship between the two. That is the performance of the VIX index may not be replicated by buying or selling securities.

Short positions in the two long VIX ETN’s would result in positive performance. The VXX returns 16.85% and the VXZ returned only 6.29%. In all three comparisons of big moves in the VIX index, a short position in the VXX has outperformed a short position in the VXZ. This is to be expected as the VXX has exposure to shorter dated VIX futures contracts that are more sensitive to a change in the VIX index.

So what have we learned from analyzing past performance of VIX trading vehicles in times of high volatility? First, even though trading vehicles that derive their value from the VIX do not move in the same magnitude as the underlying index, they do move in the same direction. Also, different trading methods move in the same direction, based on the same drop in the VIX they all have very different final results. Finally, short positions in the VIX futures did not profit nearly as much as would be expected by looking at the drop in the index, the performance shorting these contracts was pretty consistent relative to the index weakness.

Russell Rhoads

optionpit.com

http://expiringmonthly.com/