One thing I like about futures and options is those markets offer additional ways to make a good trade, even when we have no particularly strong view on the direction of the market. With futures, we can make bets not just on the outright contract price, but the slope and shape of the calendar curve.
Today we look at the VIX (VXX) futures curve. We ordinarily would expect this curve to be in contango, not because of any cost of carry like a traditional physical commodity, but because an event that causes extreme volatility this month (like a natural disaster, war, terrorism, etc.) is likely to spill over into next month. Add to that any new events that occur next month. So, normally, volatility is more expensive the further you go into the future. Of course, there are times, typically right after an extreme event, when the market expects volatility to dissipate and therefore we see backwardation in the curve.
So, let's find a trade idea. We can plot the contact prices for each month and see if it conforms to my expectations. (I used the Bloomberg CCRV <GO> function for this.) We see it is in contango, except for one month, December 2012. The market is saying that the S&P 500 should be more volatile in November and January than in December. While it may reasonable to expect a year-end holiday lull, that is a very precise forecast for a temporary dip in volatility seven months away.
To bet against the December lull, we can buy December VIX contracts and sell one or more of the other months. We could sell November only, or we could split our sales against November and January. Since these contracts become less liquid the further we move out the curve, we will have to be patient putting this position on and waiting for it to evolve, especially if we choose to sell January. The Nov-Dec-Jan butterfly is the purest way to express our view about December, but we gain some liquidity if we sell November only. We could express our view in alternative ways as well. For example, would could sell October against December at approximately even volatility. That calendar spread has recently traded between positive and negative 15 cents and now looks like it is starting to move into the 25 cent area. Compare that to the midrange two-month calendar spreads net around $2.00. That spread also works best if volatility falls or stays roughly constant.
The risk in this type of trade is we experience an extreme market event that causes volatility to spike and the curve to go into backwardation. A butterfly spread would help mitigate against that risk, and being several months away from spot also helps damp market reaction. As an example, look at what would have happened to this type of trade in a bad scenario, such as the result of the May 6th 2010 "flash crash," depicted in the chart below.
Comparing the futures curve from the day before the flash crash (lower) to the close on the day of the flash crash (upper), we see that the spot VIX rallied from 24.91 to close at 32.80, and the front contract gained more than 5 points. However, if we had a short October position against November, in an equal number of contracts, we would have broken even on the day.
In essence, the today's December contract could gain about $0.65 against its neighbors or possibly 2 points against the October contract over the next couple of months without a whole lot of risk. Look to put it on when the S&P 500 is down on the day and the curve flattens a little to get a great entry point.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
