Today’s Jump in the VIX Futures Term Structure

When we see a large VIX spike, I invariably see a similar spike in emails with the theme of “Are we there yet?” Naturally, investors want to know how far the VIX is likely to move, when the markets should be expected to reverse and whether they should trade with the rising VIX or against it.

These are all great questions and for the most part, the answers are not as simple as just plugging the current data into a spreadsheet and having it spit out something like another 2.16 points, 1.5 days and ‘start fading the VIX 45 minutes after tomorrow’s open…’

What is possible is to look at prior VIX spikes and understand some statistical tendencies. The more difficult part is to put VIX spikes in an overall context and answer questions such as whether the market had been trending up or trending down, whether the proximate cause of the selloff in stocks is geopolitical, macroeconomic, technical or whatever. (See Forces Acting on the VIX for a laundry list of potential influences on volatility.) Combining the art of context with the science of statistics, we are now on the doorstep of the dark art of forecasting volatility.

The best place to start when evaluating the current state of volatility is with the market’s expectations of future volatility and the best place to gather this information is in the form of VIX futures (or perhaps SPX implied volatility). The graphic below shows how market expectations of future VIX values have changed from yesterday’s close (dashed blue line) to about 1 hour and 40 minutes into today’s session (solid red line.) While it certainly has felt like a huge move, with the VIX up 17%, the 10.1% move in the front month (August) VIX futures still leaves us well below where investors believe the VIX will be in September and October.

The other important point to note, particularly if you are relatively new to VIX futures and want to understand them in the context of how they influence the VIX ETNs (VXX, XXV and VXZ), is that the VIX futures moves are not proportional. In fact, the front month typically moves much farther in percentage terms than the second month does, the second month moves more than the third month, and so on. One look at VIX Futures: The One Picture to Remember should drive that point home rather forcefully. The reason why front month futures are more sensitive to changes in the VIX than back month futures has to do with mean reversion. Simply stated, any one day movement in the VIX should do very little to change one’s view of where the VIX will be six or seven months out. On the other hand, with VIX August futures and options set to expire one week from today, the ripple effect from today’s action is certain to be felt in terms of how investors are looking at stocks – and volatility – one week from now. This also gets at the essence of the difference between VXX and VXZ as well as the slope of the term structure (whether it be in contango or backwardation) in the front months versus the back months.

For more on related subjects, readers are encouraged to check out:

• Weekly Options Gain Momentum
• Forces Acting on the VIX
• A Conceptual Framework for Volatility Events
• Availability Bias and Disaster Imprinting
• More Volatility + Less Fear = Lower VIX?
• VIX Term Structure Changes Since November 20th
• VIX Futures: The One Picture You Need to Remember

[source: FutureSource.com]

Disclosure(s): short VIX and VXX at time of writing

Dubai Debt Concerns Trigger Spikes in Foreign Volatility Indices

With the S&P 500 and NASDAQ futures both down approximately 3% as I type this on concerns about the ability of Dubai World to repay some $59 billion in debt, tomorrow’s half day session is likely to be ugly and volatile. In a vacuum, this type of event would be concerning, but not likely to cause panic. With the emotional scars of the financial crisis still looming large in the memories of investors (i.e., Availability Bias and Disaster Imprinting), I would not be surprised to see an overreaction in the markets tomorrow.

On average, a 3% drop in the SPX yields a spike of about 12.6% in the VIX. Yesterday the Dow Jones STOXX 50 index of European companies fell 3.36%, yet the VSTOXX, (the corresponding volatility index, which is similar to a pan-European VIX) spiked 28.16%.

For the record, a 28.16% increase in the VIX would put it at 26.25.

With the shortened trading session tomorrow providing below average liquidity and investors having an entire weekend to fret about possible contagion or additional cockroaches suddenly appearing, I am predicting a wild ride.

My general approach to events like this one is to try to fade the VIX spike as it shows signs of having topped, but there are frequently multiple spikes to contend with in these types of scenarios.

I will do the best I can to provide some commentary and analysis as the events of the Dubai World debt problems unfold.

In the interim, readers who are interested in previous posts on related subjects, readers are encouraged to check out:

• A Conceptual Framework for Volatility Events
• Forces Acting on the VIX
• Availability Bias and Disaster Imprinting
• VIX Data to Support Availability Bias and Disaster Imprinting Hypothesis

Forces Acting on the VIX

I have received quite a few requests to comment on the recent falling VIX, which stands at 23.23 as I write this, as well as the VIX:VXV ratio, how far I expect the current bull leg to run, etc.

I will get to most of this over the weekend (and newsletter subscribers will invariably get a much more detailed sense of my thinking), but I thought this might be a good time to put up a graphic that attempts to capture some of the many forces that act on the VIX. Going forward, I believe having a framework to refer to when talking about the VIX might help ground some of the dialogue.

When all is said and done, the VIX reflects supply and demand for options on the S&P 500 index. The factors that affect movements in the VIX from day to day or week to week, however, are always in flux. The graphic below is the result of a brain dump I did this morning in an effort to put some of these factors onto a single page. I started in on grouping the forces that act on the VIX and using some arrows to indicate relationships between the various factors, etc., but this clearly requires a little more soak time before it will look like a finished product. For that reason, I thought I might post this graphic here and ask for reader feedback.

[Note that the relative positions of the shapes on the VIX axis are not necessarily indicative of the potential effect they might have on the VIX. At first I wanted the graphic to encapsulate each factor on a relative importance scale and yet also have grouping and arrows that helped to described the relationships across factors. I think this might have been a little too much wishful thinking in just two dimensions, so the graphic below has some of the relative importance and some of the relationships, but is far from the last word on the subject.]