Market Breadth Primer: The VIX Phenomenon
Majority of option traders who understand options tend to focus on the mathematics of VIX while the chart traders who trade S&P 500 Index futures tend to focus on the chart patterns and confirmation signals from VIX.
Who is right?
Who understand VIX better?
Here is a short piece on the VIX Phenomenon that no one in the financial blog sphere ever talked about. Maybe learning something about VIX from a completely different perspective all together can help us answer the questions above.
CBOE Volatility Index
VIX is the symbol for CBOE Volatility Index. I am not going to waste time here discussing what it is. You can find out more about the index at wikipedia. Links are provided below for further reading.
The one single most important key about the VIX value itself is that it is an annualized rate of the expected variance of the S&P 500 based on the nearest traded strikes of the SPX options.
Thus the value of VIX cannot be directly translated into the expected volatility until you plug the value into a formula to get the estimated standard deviation of the period you are interested in. The calculation method is quite simple. There is an example in wikipedia illustrating the method so I am not going to repeat that here.
Let’s dive in to the heart of the issue.
The VIX Phenomenon
Following is a table of VIX levels with the corresponding implied 30 days percentage change and the translated absolute value at 100 points interval based on S&P 500 price levels. If you do not understand what I mean in the first sentence, it is okay. Let’s look at the table first and I will explain right after.
When VIX is at 13, the 30 days estimated volatility is 3.75% and that translate into 45 S&P points if S&P is trading at 1200. If S&P is trading around 1800, that translates into 67.55 points. These absolute values carry a lot of weight. They are there telling you that if S&P is trading at 1200 with VIX at 13, there is a 65% chance S&P is going to stay within +/- 45 points from 1200. If you trade options, you must know this to formulate your strategy properly.
So many numbers in a huge table. What’s the point?
I have computed the average value of VIX since 1997 at each 100 point interval on the S&P 500. It is the 2nd line from the top of the table. For example, when S&P was trading at 1200 +/- 50 points, the average value of VIX was 20.4 since 1997.
Looking at the numbers themselves it is not easy to draw a connection. So, I have highlighted these average VIX levels in yellow corresponding to the respective S&P price levels. Now, they are very interesting.
The higher volatility readings at 1200 to 1500 was mainly due to the steep decline and spike values in VIX during the last 2 market crashes. But that does not change the fact that there is a general trend of decline in VIX level relative to the increase in S&P 500 price levels.
Why is it the case?
Keep It Simple Stupid
The clue is the near constant range of the absolute values translated from these standard deviations.
From the 1600 to 1800 column, the absolute value is within a constant range at 65 to 75 S&P 500 points.
If the extreme spikes (value greater than 40) are removed from the VIX historical readings, this constant range will be the same for the 1300 to 1500 columns too.
Traders who trade S&P 500 options, be that used for hedging or pure speculation, do not really depend their final trading decisions on their option models. When things boil down to money making, these traders are really looking for the absolute potential in their trades and adjust the risk they are going to take accordingly. After all, profit and losses are measured in S&P points, not the Greeks in the option price models.
In short, options on index are traded off the absolute value of the indices because human do not trade with unlimited capital and fictitious percentage measures. Humans behave the same way no matter S&P trading at 1200 or 1800. This phenomenon also indirectly shows that majority of volatility models failed to capture the essence of human trading behaviour.
It is now easy to answer the questions I raised in the beginning of this article.
The option traders more talented in math are blinded by their superior math skills and ignored the fundamental principle of any market. That is, traders are bounded by their profit making objectives and risk profiles. Thus, no matter what they think, VIX has the characteristics of a tradable instrument and it shows in its charts.
The traders who are good at chart reading can often pick up the interesting behaviour of VIX like how it loves to hold a particular level for a long period of time. That is excellent detective work but the lack of understanding the nature of VIX makes it difficult to figure out the reasons behind such phenomenon. In turn, the inference from the observations may not be as accurate as one likes it to be.
Both sides have something to learn from each other.