In Chapter 9 of Trading VIX Derivatives I discuss VIX as a stock market indicator. The relationship between VIX and VIX futures is discussed and to smooth futures data I created what I call the Modified VIX Futures contract. The motivation behind creation of this Modified VIX Futures contract was to have a consistent futures price to compare to VIX. Simply put, the Modified VIX Futures contract takes a time weighted average of the front two month futures contracts to create a single futures price to compare to the VIX index. As focus shifts from the near month to the second month in trading volume so does the emphasis of influence from the price of the second month futures contract. On the Friday before expiration I will retired the front month and start using the next two expiration month contracts. This Friday will be the last day that January VIX Futures feed into the calculation as January VIX Futures and Options expire next week.
A use that had not occurred to me when I was writing the book was that the Modified Future may be a method of determining when near term VIX futures are in contango or backwardation relative to the VIX index. These two terms, contango and backwardation, refer to the shape of the graph that may be created by charting and underlying market and futures prices according to time left to expiration.
When the market goes into what is known as backwardation the biggest impact is often on the nearer dated contracts. The table below shows the number of days the Modified Future has been at a discount to the spot index for each year since 2007. I always use January 2, 2007 as a starting date for any VIX Futures related back test as volume was sufficient at that time to lead credibility to the data I use for the study.
Trading Days Contango Backwardation % Contango % Backwardation
2007 251 177 74 70.52% 29.48%
2008 253 143 110 56.52% 43.48%
2009 252 202 50 80.16% 19.84%
2010 252 223 29 88.49% 11.51%
2011 252 178 74 70.63% 29.37%
2012 250 248 2 99.20% 0.80%
Total 1510 1171 339 77.55% 22.45%
What stands out dramatically on this table is 2012. There were only two trading days where VIX closed at a premium to the Modified Future. In fact those two days were December 27th and 28th when the S&P 500 was under pressure based on fiscal cliff fears going into the end of the year. Backwardation is usually the result of a spike in VIX. Volatility will revert to a mean or average over time. When VIX moves up quickly the futures markets subsequently price in a drop. With one small exception a spike in VIX just did not occur in 2012. Also, the high low range for VIX in 2012 was the narrowest range since 2012. The table below compares shows the high low range for VIX by year from 2007 to 2012 –
2007 2008 2009 2010 2011 2012
High 31.09 80.86 56.65 45.79 48.00 26.66
Low 9.89 16.30 19.47 15.45 14.62 13.45
Range 214% 396% 191% 196% 228% 98%
Average 17.54 32.69 31.48 22.55 24.20 17.80
The high low range for 2012 was less than 100% which is a much narrower range that any other year since 2007. I believe comparing the yearly ranges and average VIX prices is more significant than looking at year to year change VIX. VIX is going to oscillate more than trend so comparing the close at the end of 2011 to the close at the end of 2012 does not tell me much. If December 30th had been the end of 2012 instead of December 31st VIX would have been down 2% in 2012 instead of down 22%. Again, I think range and average VIX tells us more. This table tells me is what the nature of VIX was in 2012 in context to other years. VIX was low and stayed pretty low for most of the year.
Finally, there are two things VIX traders use as a rule of thumb. One is the reversion of VIX that I mentioned above. The second, that when VIX and market volatility has been calm eventually a storm shows up and VIX spikes accordingly. The longer the quiet, the closer the big VIX move may be. Seeing 2012 as a quiet year makes me wonder what 2013 will bring. It’s been quiet, maybe too quiet.
What % Allocation to VIX Futures and Options for Portfolio Diversification?
During times when the CBOE Volatility Index® (VIX®) is at relatively low levels, we often receive investor questions such as – how much of an allocation might I make to VIX futures and options in order to try to diversify my portfolio?
RELATIVELY LOW RECENT VALUES FOR VIX
In 2012 the average daily closing value of the VIX was 17.8, its lowest such value since 2007.
PAPER BY UNIVERSITY OF MASSACHUSETTS
In trying to answer the question re: allocation to VIX, one could explore the 2009 paper by the University of Massachusetts “
For a traditional portfolio of stocks, bonds and alternatives during the five-month time period from August through December 2008, the following are three ways in which long volatility exposure was added, and the results are presented for the 5-month period studied:
(1) Using a 10% allocation to long near-term CBOE VIX futures–
- Total returns were improved by 15.7 percentage points (improvement to -4.0% from -19.7%)
- Standard deviation was reduced by about one-third (to 16.3% from 25.3%)
(2) Using a 3% allocation to long at-the-money one-month CBOE VIX calls, total returns were increased to +20.8% from -19.7%
(3) Using a 3% allocation to long 25%-out-the-money one-month CBOE VIX calls, period returns increased to +97.2% from -19.7%
The paper concludes by noting that “…investable VIX products could have been used to provide some much-needed diversification during the 2008 financial crisis.” A link to the paper is available at www.cboe.com/vix. Two of the figures from the